Podcasts about r d

Loral Langemeier is the founder and CEO of several alternative asset companies, with decades of experience in finance, mentoring, real estate investing, business development, and oil and gas. She is a six-time New York Times bestselling author, host of Real Money Talks, and a highly sought-after money management expert who has built multiple successful businesses. Go to www.askloral.com to get Lorla's book for FREE as well as 2 free tickets to her webinar.   Here's some of the topics we covered: Loral Langemeier's journey from Nebraska farm girl to millionaire Building wealth through real estate, oil, and alternative investments Why traditional financial services fail to integrate wealth strategies Proactive tax planning, corporate structures, and asset protection Hidden tax strategies like R&D credits and cost segregation Creating multiple income streams and building a lasting legacy Educating the next generation to protect and sustain family wealth   To find out more about partnering or investing in a multifamily deal: Text Partner to 72345 or email Partner@RodKhleif.com    For more about Rod and his real estate investing journey go to www.rodkhleif.com   Please Review and Subscribe  

Bryan Murphy, CEO of Smartling, confronts a $40 billion industry stuck in the slow lane—the world of translation. Most companies still handle translations like it's 1999: manual, expensive, and painfully slow. Bryan saw AI as the game changer that could rewrite the rules, but integrating it wasn't a walk in the park. He shares how Smartling harnessed AI to not just cut costs and speed up translation but to finally boost quality close to human-level precision without losing control over brand voice or nuance. Yet, making this leap meant upheaval: reorganizing teams, hiring AI experts, and establishing ruthless R&D discipline to separate winning ideas from distractions. Bryan also shares: - Why a $40 billion translation industry was ripe for disruption - How AI helped Smartling move from faster and cheaper to dramatically better - What it took to bring AI into a traditional business without sacrificing quality or brand voice - Why hiring the right AI talent was critical to making the shift - How ruthless R&D discipline helps separate breakthrough ideas from distractions - Why every AI initiative needs a clear connection to customer outcomes Chapters 00:00 - Introduction to Bryan Murphy and Defining the Translation Challenge 02:30 - The Hidden $40B Translation Market and Its Untapped Potential 06:00 - Early AI in Translation: Faster and Cheaper, But Not Yet Better 09:20 - Human-in-the-Loop AI: Boosting Translator Productivity Tenfold 13:00 - Unlocking Market Expansion Through Improved SEO and Digital Footprint 15:00 - The Moment of Truth: Recognizing GPT's Impact and Rolling Out Rapid Innovation 18:00 - Overcoming Organizational Challenges: From Excitement to Structured Execution 22:00 - The Discipline of "Customer-First" in AI Development and Roadmapping 24:00 - Leadership Lessons: Listening Without Losing Vision Amidst Painful Change 26:00 - Winning Customer Trust: Betting On Proofs of Concept Against Skeptics 28:00 - Personal Insights: Favorite Leadership Books and the Role of Intellectual Curiosity 33:00 - Wrap-up and Invitation to Follow Smartling's AI-Empowered Evolution Try Smartling: smartling.com

The fastest way to find out if your strategy holds water? Put it in the market.  For more than a decade, Cheeky Cocktails has built a following with premium syrups and juices designed to elevate at-home cocktails.  Now, founder and CEO April Wachtel is pushing the brand into a first-of-its-kind functional tonic, and shares how real-world feedback is reshaping everything from where the product belongs on the shelf to how she's thinking about Cheeky's next chapter. For CPG founders, it's a compelling case for embracing iteration rather than waiting for perfection. Show notes: 0:20: April Wachtel, Founder & CEO, Cheeky Cocktails – April discusses her unconventional path from working in restaurants as a teenager to becoming a craft cocktail bartender, spirits brand ambassador and consultant before launching her own company. She reflects on the evolution from her original brand, Swig and Swallow and how she ultimately relaunched the business just as the COVID-19 pandemic accelerated at-home cocktail culture. April explains how her experience as a bartender, cocktail instructor and consumer shaped Cheeky's portfolio of premium syrups and juices, as well as its newest product, Cheeky Lime Tonic, which she describes as a functional tonic with fiber and 40–70% less sugar than traditional tonics. April details the extensive R&D and consumer feedback that went into developing the tonic and explains why she believes entrepreneurs should get products into the market quickly to test their assumptions and learn from real-world reactions. She also discusses Cheeky's evolving retail strategy and the challenge of building a company around two distinct product lines and the possibility that the tonic could eventually become the primary focus of Cheeky. April  also shares how intuition, personal experience and data have each influenced her entrepreneurial decisions, emphasizing the importance of following instinct while validating ideas with market research and consumer feedback. Brands in this episode: Cheeky Cocktails, Poppi, Olipop, Culture Pop

Paul Lunter didn't start out trying to build a wellness empire — he was in construction, watched someone in pain step in and out of the shower for relief, and built a machine instead. Thirty-seven years later, that machine (HydroMassage) has grown into WellnessSpace Brands: five product lines, 400 million sessions delivered, and installations in 50+ countries at Planet Fitness, Life Time, Gold's Gym, and beyond. Lunter walks Eric Malzone through the company's medical-market origins, the R&D behind machines like the PolarWave Dry Plunge and the company's newest RedZone sauna which uses no wood, and why he thinks recovery is a permanent industry shift rather than another fitness fad. Key Takeaways

Larry Patterson went in for a routine hernia repair. He came out, eventually, with several inches of his bowel removed. Timothy O'Brien, board-certified trial attorney and shareholder at Levin Papantonio, just won $88 million for Patterson and his wife Tammy against Covidien, a Medtronic company. It is the largest compensatory award in fifteen years of mesh litigation, and the first of roughly 2,500 cases to reach a jury. The mesh, Symbotex, sits right against your intestines behind a collagen barrier meant to last 30 days. Covidien's own internal testing showed it was gone in under seven. They knew in 2003. O'Brien walks Gregg through the evidence that turned the trial, where sales staff swore to doctors the barrier lasted a month while R&D said otherwise in writing. He also maps the hernia mesh lawsuit landscape, from the Bard settlement covering 33,000 claimants to why Medtronic Covidien is the last major manufacturer still fighting, and what a bellwether verdict actually means for the thousands of people still waiting on their own cases   Join Gregg and Timothy O'Brien on Climate Change Environment Science & the Law as they explore:   What You'll Learn Why a generation of surgeons stopped doing tissue-to-tissue repair, and what that traded away How Symbotex's protective barrier was supposed to work, and how fast it actually disappeared The 2003 document that locked in the depositions and became the case's spine Why sales and R&D were telling two completely different stories about the same product What a bellwether trial is, and why the first one carries the most weight Where the Bard, Johnson & Johnson, and Medtronic Covidien litigations stand right now Why about 100 new Covidien cases are still being filed every month What Covidien's planned appeal means for people still waiting How to find out if you or someone you love has a case   Time Stamps 0:00 – "They are lying": the sales pitch vs. the truth 0:48 – Meet attorney Timothy O'Brien 1:02 – The hernia mesh "safety feature" that failed 2:06 – Why hernia mesh exists in the first place 4:13 – The Patterson case: what happened to the plaintiff 5:41 – The depositions that exposed Covidien 7:46 – 2003: when they allegedly knew the truth 9:33 – The 3 major mesh manufacturers, explained 11:59 – What a "bellwether trial" means for your case 14:07 – Will Covidien settle or appeal? 17:40 – How to contact Timothy O'Brien about your case   Timothy O'Brien is a board-certified trial attorney and shareholder at Levin Papantonio, where he has served as lead or co-lead counsel in mass tort litigation against major hernia mesh manufacturers, including Bard, Johnson & Johnson, and Medtronic Covidien. He recently helped secure an $88 million verdict in the first bellwether trial of the Covidien hernia mesh multidistrict litigation.    Contact / Follow Timothy O'Brien:  Firm: Levin Papantonio  Website: https://levinlaw.com     Want more conversations that cut through the noise on science, climate, and the issues shaping our future? Subscribe to Climate Change Environment Science & the Law with Gregg Goldfarb for new episodes every week.   

What happens when a founder decides to pull a product off 400 store shelves not because it failed, but because scaling the wrong economics is a fast track to killing a company. In this masterclass on CPG entrepreneurship, Omar Atia, Co-Founder & CEO of ZeroCarb LYFE, joins host Rose Hamilton, CEO of Compass Rose Ventures, to break down the mechanics of building a resilient, high-value enterprise.  Omar traces his evolution from corporate R&D to startup leadership, delivering a candid blueprint for navigating the hidden traps of rapid expansion, commercializing novel manufacturing processes, and making disciplined financial trade-offs. Key Takeaways & Business Frameworks: * Unit Economics Over Top-Line Growth: Why exiting 400 Sprouts locations was the smartest move for long-term enterprise value, proving that revenue without margin is just noise. * Bridging R&D and Commercial Scale: The operational reality of transforming a four-ingredient, kitchen-table innovation into a scalable, patented manufacturing system. * Positioning for Market Realities: How tracking customer behavior led ZeroCarb LYFE to pivot its core value proposition from "low-carb" to "highest protein-to-calorie ratio," proving why messaging must evolve alongside the market. * Omnichannel Strategy Demystified: A tactical breakdown of how to use direct-to-consumer for immediate feedback loops, food service for margin stability, and selective retail for broader reach. * Capitalizing on Structural Consumer Shifts: How to build product strategies around macro health shifts, including the rise of GLP-1 medications, by prioritizing high-protein, calorie-efficient nutrition. Whether you are scaling a CPG brand, refining your go-to-market strategy, or evaluating channel profitability, this interview serves as a practical guide to operational discipline and sustainable growth. For more on ZeroCarb LYFE visit: https://zerocarblyfe.com/ If you enjoyed this episode, please leave The Story of a Brand Show a rating and review.  Plus, don't forget to follow us on Apple and Spotify.  Your support helps us bring you more content like this!

Virtual cell models promise faster, cheaper early-stage drug discovery. However, the industry still lacks a shared way to judge which of these models can actually be trusted on a given problem.   In this episode, Kristóf Szalay, CTO and Co-Founder of Turbine, and Gerold Csendes, Scientist at Turbine, set out what virtual cell models can and can't do today, in conversation with host Marilie Fouché. They cover why benchmarking remains fragmented across the field, how pharma teams build confidence in a model before trusting it with real R&D decisions, and how compressing the feedback loop between experiments can cut months out of the discovery process.   This episode is sponsored by Turbine.   Emerj works with a select group of AI vendors to reach Fortune 500 decision makers through research, media, and direct access. If you want to be considered, download our media kit at emerj.com/AD1

Today our hosts are joined by Isaac Nelson to talk about the development of Beam Splitter. Isaac is the other half of OBNE's R&D department, and an actor of some renown, recognizable for characters such as Old Ike and whatever we call the corporate guy who drinks the Sunlight tincture. Isaac gives nod to his object (a blue strat) and describes himself as not a particularly curious gear guy, while also giving insight into the inspiration behind his triple tracking multi-distortion mono stereo trereo embiggening behemoth, Beam Splitter. Buy some Old Blood: https://oldbloodnoise.com/ Join the conversation in Discord: https://discord.com/invite/PhpA5MbN5u Follow us all on the socials: @danfromdsf, @andyothling, @oldbloodnoise Leave us a voicemail at 505-633-4647!

Canada's venture capital market is showing signs of life, but the recovery comes with a catch: more money is flowing into fewer companies. Matt Cohen and John Ruffolo unpack what that means for Canadian founders, why promising tech companies keep getting acquired before they reach scale, and whether Canada is at risk of becoming the world's R&D shop instead of building lasting global champions.They also dig into Canada's $2.3 billion Telesat sovereignty bet, the growing backlash against AI data centers over energy and water use, and Mark Zuckerberg's push for a more decentralized AI future. John agrees with parts of Zuckerberg's argument, but questions whether the Meta founder is really the right person to lecture the world about distributing power.Finally, Matt and John break down Ontario Teachers' massive SpaceX win and what it could mean for how Canadian pension funds think about venture capital. If you want a sharper read on where Canadian tech, AI infrastructure, venture capital, and economic sovereignty are heading next, this episode is a must-listen.–A big thanks to our sponsor, Moomoo CanadaThis is the kind of tooling that used to live on a Bloomberg terminal, but now it is on your phone, just a few taps away. They offer real-time data, full options chains, and an AI assistant that actually explains trading strategies.Moomoo is the perfect place for people who want to take their money seriously. Open an account today at moomoo.caCanada's VC Rebound Comes With a Catch (01:53)Canadian venture funding is rising, but deal volume keeps falling as more capital concentrates around a smaller group of companies.John explains why this “flight to quality” could create huge winners while making life much harder for strong startups outside the hottest deals.Canada's $2.3B Sovereignty Bet (06:15)Canada's major investment in Telesat and secure satellite communications is about much more than competing with Starlink.John argues that defence and other sensitive infrastructure cannot depend entirely on foreign providers, making sovereign technology a strategic necessity.The AI Data Center Backlash Is Getting Political (09:19)AI data centers are running into growing resistance over electricity demand, water use, and the strain they place on local infrastructure.John believes the issue could become much more political in Canada if rising AI demand starts pushing up energy costs for everyday consumers.Zuckerberg Wants to Decentralize AI Power. Really? (14:48)Mark Zuckerberg is pushing a vision of personal superintelligence that puts powerful AI tools into the hands of individuals and small businesses.John agrees with some of the open-source philosophy, but calls out the contradiction of hearing a message about decentralizing power from the founder of Meta.Can One SpaceX Bet Change How Pension Funds Think? (18:55)Ontario Teachers' SpaceX investment shows how a single exceptional venture bet can have a meaningful impact even inside a massive pension portfolio.John explains why pension funds should remain disciplined, but says wins like SpaceX and Shopify prove that venture capital absolutely can move the dial.Connect with John Ruffolo on LinkedIn: https://ca.linkedin.com/in/joruffoloConnect with Matt Cohen on LinkedIn: https://ca.linkedin.com/in/matt-cohen1Visit the Ripple Ventures website: https://www.rippleventures.com/ This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit tanktalks.substack.com

Apple's latest financial results drive a discussion about record iPhone, Mac, Services, subscription, and R&D numbers, along with the market's negative reaction and stock volatility. The MacVoices Live! panel also examines leadership changes under new CEO John Ternus, including the return of a retired engineering executive, before debating Apple's planned clipboard sharing between iPhones and Windows PCs, its implications for users, and whether it introduces meaningful security concerns.  Panel members include: Marty Jencius, David Ginsburg, Jim Rea, Chuck Joiner, Eric Bolden, Web Bixby, and Brian Flanigan-Arthurs. This edition of MacVoices is supported by MacVoices Magazine, our free magazine on Flipboard. Updated daily with the best articles on the web to help you do more with your Apple gear and adjacent tech, access MacVoices Magazine content on Flipboard, on the web, or in your favorite RSS reader. Show Notes: Chapters: 00:00 Introduction, Riverside updates, and panel introductions04:20 Marty celebrates his 42nd wedding anniversary07:30 Apple's quarterly earnings and Wall Street's reaction15:10 Record iPhone, Mac, Services, subscriptions, and R&D spending20:15 John Ternus' influence on the MacBook Neo24:30 Apple brings a retired hardware executive back to the company31:40 Organizational changes under Apple's new leadership34:45 UK App Store steering rules and regulatory concerns37:15 Apple brings clipboard sharing between iPhone and Windows PCs43:15 Security, usability, and the future of cross-platform clipboard support Links: iPhone, Mac and Services revenue set new June quarter records https://appleworld.today/2026/07/iphone-mac-and-services-revenue-set-new-june-quarter-records/   Apple spent a record $11.73 billion in R&D during Q3 2026 – 9to5Mac https://9to5mac.com/2026/07/30/apple-spent-a-record-11-73-billion-in-rd-during-q3-2026/   Apple just hit 1.5 billion paid subscriptions from its user base – 9to5Mac https://9to5mac.com/2026/07/30/apple-just-hit-1-5-billion-paid-subscriptions-from-its-user-base/   Ternus's MacBook Neo pushes Mac sales to unexpected Q3 heights https://appleinsider.com/articles/26/07/30/ternuss-macbook-neo-pushes-mac-sales-to-unexpected-q3-heights?utm_source=rss   Team Ternus drags retired vice president of hardware engineering back into Apple https://appleinsider.com/articles/26/08/03/team-ternus-drags-retired-vice-president-of-hardware-engineering-back-into-apple?utm_source=rss   Apple Says UK App Store Steering Rules Would Be ‘Highly Intrusive' https://www.macrumors.com/2026/07/29/app-store-uk-rules-highly-intrusive/   Apple is finally making copy-paste work between iPhones and PCs https://www.macworld.com/article/3205820/apple-is-finally-making-copy-paste-work-between-iphones-and-pcs.html Guests: Get detailed bios and contact information about for the panel on the MacVoices Live! Panel page on our web site:https://macvoices.com/macvoiceslive/macvoices-live-panel/ Support:      Become a MacVoices Patron on Patreon     http://patreon.com/macvoices      Enjoy this episode? Make a one-time donation with PayPal Connect:      Web:     http://macvoices.com      Twitter:     http://www.twitter.com/chuckjoiner     http://www.twitter.com/macvoices      Mastodon:     https://mastodon.cloud/@chuckjoiner      Facebook:     http://www.facebook.com/chuck.joiner      MacVoices Page on Facebook:     http://www.facebook.com/macvoices/      MacVoices Group on Facebook:     http://www.facebook.com/groups/macvoice      LinkedIn:     https://www.linkedin.com/in/chuckjoiner/      Instagram:     https://www.instagram.com/chuckjoiner/ Subscribe:      Audio in iTunes     Video in iTunes      Subscribe manually via iTunes or any podcatcher:      Audio: http://www.macvoices.com/rss/macvoicesrss      Video: http://www.macvoices.com/rss/macvoicesvideorss

Apple's latest financial results drive a discussion about record iPhone, Mac, Services, subscription, and R&D numbers, along with the market's negative reaction and stock volatility. The MacVoices Live! panel also examines leadership changes under new CEO John Ternus, including the return of a retired engineering executive, before debating Apple's planned clipboard sharing between iPhones and Windows PCs, its implications for users, and whether it introduces meaningful security concerns.  Panel members include: Marty Jencius, David Ginsburg, Jim Rea, Chuck Joiner, Eric Bolden, Web Bixby, and Brian Flanigan-Arthurs. This edition of MacVoices is supported by MacVoices Magazine, our free magazine on Flipboard. Updated daily with the best articles on the web to help you do more with your Apple gear and adjacent tech, access MacVoices Magazine content on Flipboard, on the web, or in your favorite RSS reader. Show Notes: Chapters: 00:00 Introduction, Riverside updates, and panel introductions 04:20 Marty celebrates his 42nd wedding anniversary 07:30 Apple's quarterly earnings and Wall Street's reaction 15:10 Record iPhone, Mac, Services, subscriptions, and R&D spending 20:15 John Ternus' influence on the MacBook Neo 24:30 Apple brings a retired hardware executive back to the company 31:40 Organizational changes under Apple's new leadership 34:45 UK App Store steering rules and regulatory concerns 37:15 Apple brings clipboard sharing between iPhone and Windows PCs 43:15 Security, usability, and the future of cross-platform clipboard support Links: iPhone, Mac and Services revenue set new June quarter records https://appleworld.today/2026/07/iphone-mac-and-services-revenue-set-new-june-quarter-records/   Apple spent a record $11.73 billion in R&D during Q3 2026 – 9to5Mac https://9to5mac.com/2026/07/30/apple-spent-a-record-11-73-billion-in-rd-during-q3-2026/   Apple just hit 1.5 billion paid subscriptions from its user base – 9to5Mac https://9to5mac.com/2026/07/30/apple-just-hit-1-5-billion-paid-subscriptions-from-its-user-base/   Ternus's MacBook Neo pushes Mac sales to unexpected Q3 heights https://appleinsider.com/articles/26/07/30/ternuss-macbook-neo-pushes-mac-sales-to-unexpected-q3-heights?utm_source=rss   Team Ternus drags retired vice president of hardware engineering back into Apple https://appleinsider.com/articles/26/08/03/team-ternus-drags-retired-vice-president-of-hardware-engineering-back-into-apple?utm_source=rss   Apple Says UK App Store Steering Rules Would Be 'Highly Intrusive' https://www.macrumors.com/2026/07/29/app-store-uk-rules-highly-intrusive/   Apple is finally making copy-paste work between iPhones and PCs https://www.macworld.com/article/3205820/apple-is-finally-making-copy-paste-work-between-iphones-and-pcs.html Guests: Get detailed bios and contact information about for the panel on the MacVoices Live! Panel page on our web site: https://macvoices.com/macvoiceslive/macvoices-live-panel/ Support:      Become a MacVoices Patron on Patreon      http://patreon.com/macvoices      Enjoy this episode? Make a one-time donation with PayPal Connect:      Web:      http://macvoices.com      Twitter:      http://www.twitter.com/chuckjoiner      http://www.twitter.com/macvoices      Mastodon:      https://mastodon.cloud/@chuckjoiner      Facebook:      http://www.facebook.com/chuck.joiner      MacVoices Page on Facebook:      http://www.facebook.com/macvoices/      MacVoices Group on Facebook:      http://www.facebook.com/groups/macvoice      LinkedIn:      https://www.linkedin.com/in/chuckjoiner/      Instagram:      https://www.instagram.com/chuckjoiner/ Subscribe:      Audio in iTunes      Video in iTunes      Subscribe manually via iTunes or any podcatcher:      Audio: http://www.macvoices.com/rss/macvoicesrss      Video: http://www.macvoices.com/rss/macvoicesvideorss Apple's latest financial results drive a lively discussion about record iPhone, Mac, Services, subscription, and R&D numbers, along with the market's negative reaction and stock decline. The panel also examines leadership changes under new CEO John Ternus, including the return of a retired engineering executive, before debating Apple's planned clipboard sharing between iPhones and Windows PCs, its implications for users, and whether it introduces meaningful security concerns. SEO Keywords Apple, iPhone, Mac, earnings, subscriptions, R&D, JohnTernus, Windows, clipboard, security, AppStore, stock, engineering, Europe, revenue Possible Episode Titles 1. Apple's Strong Quarter Meets a Wall Street Reality Check 2. Record Revenue, New Leadership, and Apple's Windows Surprise 3. Why Apple's Stock Fell Despite Record Financial Results 4. John Ternus Reshapes Apple as Windows Integration Expands 5. Apple Earnings, Executive Changes, and the Windows Clipboard Debate Chapter Markers 00:00 Introduction, Riverside updates, and panel introductions 04:20 Marty celebrates his 42nd wedding anniversary 07:30 Apple's quarterly earnings and Wall Street's reaction 15:10 Record iPhone, Mac, Services, subscriptions, and R&D spending 20:15 John Ternus' influence on the MacBook Neo 24:30 Apple brings a retired hardware executive back to the company 31:40 Organizational changes under Apple's new leadership 34:45 UK App Store steering rules and regulatory concerns 37:15 Apple brings clipboard sharing between iPhone and Windows PCs 43:15 Security, usability, and the future of cross-platform clipboard support

What drives a seasoned R&D leader to leave a major multinational and reshape digital formulation? In the latest episode of the European Coatings Podcast, editor Yeray Lopez is joined by Nick Talken, CEO and co-founder of software provider Albert Invent, to discuss the digital evolution of laboratory environments in the coatings industry.

In "How Nordian's Platform Enables Long-Haul Autonomy", Joe Lynch speaks with Co-founder and CEO of Nordian, Michael Schramm, about how Nordian enables long-haul autonomy by combining precise positioning, satellite connectivity, and edge intelligence into a single platform. About Michael Schramm Michael Schramm is Co-founder and CEO of Nordian, the positioning and connectivity platform for Physical AI, delivering centimeter-level GNSS corrections, satellite connectivity, and fleet lifecycle management to some of the largest industrial OEMs in the Americas. A serial entrepreneur with 15+ years of executive leadership, he is a founding partner of Ambush, an applied AI engineering firm; co-founder of Echo54, an advanced sensing R&D company serving US and allied government agencies; and founder of GOAT, an acquired consumer micro-mobility company. Across nearly two decades of building companies that operate in the physical world, he kept running into the same failure point: machines break when positioning and connectivity aren't engineered as one system. Nordian exists to fix that. About Nordian Nordian is the positioning and connectivity platform for Physical AI. One platform delivers centimeter-level GNSS corrections, integrated satellite connectivity, and fleet lifecycle management to industrial OEMs across transportation, agriculture, and mining. Headquartered in Austin, Texas, and deliberately launched in the hardest environments on Earth, Nordian built South America's largest PPP-RTK network, and its platform serves 80% of the region's 20 largest agricultural OEMs. Proven where networks fail and machines can't, Nordian is now expanding globally to power autonomous operations at scale. Key Takeaways: How Nordian's Platform Enables Long-Haul Autonomy In "How Nordian's Platform Enables Long-Haul Autonomy", Joe Lynch speaks with Co-founder and CEO of Nordian, Michael Schramm, about how Nordian enables long-haul autonomy by combining precise positioning, satellite connectivity, and edge intelligence into a single platform. Physical AI is a Connectivity and Processing Challenge, Not an AI Model Problem: Current AI systems are fully capable of handling autonomous navigation, but real-world physical AI is constrained by connectivity and real-time processing capabilities. Offloading critical decisions to back-end cloud servers introduces latency, which is dangerous for heavy machinery like a 25-ton autonomous truck moving at highway speeds. Edge Computing and Local Inference Eliminate Deadly Latency: To operate safely without reliance on uninterrupted internet access, 100% of mission-critical decisions must occur directly on the device using edge computing. Nordian provides the local processing capacity needed for real-time inference, allowing autonomous vehicles, drones, and heavy equipment to operate safely in "air-gapped" environments or during brief network dropouts. Centimeter-Level Positioning Replaces Imprecise Traditional GPS: Standard GPS provides meter-level accuracy, which is acceptable for route navigation but unacceptable for vehicle control, lane-level autonomous driving, precise geofencing, or row-crop agriculture. By combining satellite signals with dedicated ground reference stations to calculate real-time differential corrections, Nordian achieves centimeter-level accuracy required for absolute control. Integration Burden is the Primary Bottleneck for OEMs: Equipment manufacturers historically acted as their own integrators—trying to bolt together separate vendors for chipsets, satellite bands, cellular modems, and edge computing. Nordian abstracts this complexity by unifying precise positioning, resilient connectivity, and edge intelligence into a single plug-and-play factory-installed package with an SDK for custom software development. A Multi-Band "N+3" Connectivity Model Bridges the Connectivity Gap: Autonomy dies where cellular coverage fails, particularly across the 71% of U.S. roadways located in rural environments. Nordian solves the connectivity gap by layering cellular networks, L-band communications, and low Earth orbit (LEO) satellite constellations (including integrated Starlink connectivity) into a redundant system capable of rapid sub-10-second signal convergence. Agricultural Battle-Testing Translates Directly to Transportation and Logistics: Before expanding into long-haul trucking and yard logistics, Nordian proved its system in South America's harsh agricultural environments, building a massive reference station network across Brazil and Argentina. This foundation enabled them to capture 80% of the top 20 agricultural OEMs in the region—proving the technology where infrastructure is non-existent and atmospheric interference (scintillation) is severe. Autonomous Technology Target Long-Haul Workloads to Improve Quality of Life: Autonomous technology is positioned to address structural labor shortages by replacing high-turnover, long-haul routes (where drivers are away from home for weeks) with fully autonomous systems or human-augmented modes. This shifts human operators toward last-mile and short-haul jobs, improving driver safety, operational utilization, and overall work-life balance. Learn More About How Nordian's Platform Enables Long-Haul Autonomy Michael Schramm | Linkedin Nordian | Linkedin Nordian Contact Nordian Nordian Authorized to Resell Starlink High-Speed Internet to Businesses & Enterprises. Nordian Expands High-Precision GNSS Positioning to Brazil Through Strategic Partnership with u-blox Federal News Network's Space Hour Podcast - Connecting devices out in remote regions Fierce Network - Nordian authorized to resell Starlink internet to businesses and enterprises Ending the 60% Waste: The Radical Shift Trucking Needs Right Now The Logistics of Logistics Podcast If you enjoy the podcast, please leave a positive review, subscribe, and share it with your friends and colleagues. The Logistics of Logistics Podcast: Google, Apple, Castbox, Spotify, Stitcher, PlayerFM, Tunein, Podbean, Owltail, Libsyn, Overcast Check out The Logistics of Logistics on Youtube

Glen catches up with Finovate's Greg Palmer to preview September's NYC Demo Parade and the evolution in the startup ecosystem he's noted. Also- the Fiserv Follies extends its run, and how CLARITY and credit cards suddenly became tied at the hip.   Links related to this episode:   Finovate Fall, September 9-11 in New York City: https://informaconnect.com/finovatefall/    Use code BIG20 for a 20% discount on Finovate registration: https://informaconnect.com/finovatefall/purchase/select-package/?vip_code=BIG20    Our recent interview with CUltivate AI CEO (and Finovate rookie demoer) Anthony Volpe: https://www.big-fintech.com/can-credit-unions-collaborative-superpowers-extend-to-ai/    CU Today on Sens. Lummis and Moreno's sudden addition as CCCA sponsors: https://www.cutoday.info/THE-feature/Lummis-Moreno-Join-Credit-Card-Competition-Act-As-Interchange-Threat-Grows    Fiserv's Q2 earnings and fiscal 2026 outlook, in its own words and that of its hometown paper: https://investors.fiserv.com/news-releases/news-release-details/fiserv-reports-second-quarter-2026-results  https://www.jsonline.com/story/money/business/2026/08/06/fiserv-reports-quarterly-earnings-miss-in-first-report-under-new-ceo/91194333007/      Mark your calendar to join us Wednesday August 19 at 3pm ET/Noon PT for our next CU Town Hall. Our guest speaker will be Brian Ley, whose new venture VerifyDial takes a fresh approach to scam detection by deploying a 411-style national phone line free of charge to community FIs. The Town Hall is free to attend as well, but advance registration is required. Come prepared for a lively discussion!  https://www.cutownhall.com/    Check out the Innovation Club- a curated group of credit union tech, data and strategy leaders that meets virtually each month- and twice a year in person- to extend their R&D budgets and collaborate on tangible solutions to the latest challenges. Learn more to see if you're a fit and if so, request a guest pass: https://www.big-fintech.com/innovation-club/    Follow us on LinkedIn:  https://www.linkedin.com/company/best-innovation-group/   https://www.linkedin.com/in/jbfintech/  https://www.linkedin.com/n/glensarvady/

Kristyn Carriere went from running taste panels at Cadbury and formulating for Godiva to founding her own Canadian chocolate brand. This is what happens when a real food scientist decides to build a CPG company. On this episode, Kristyn Carriere of Seven Summit Snacks joins Phil and Kenny to talk about turning deep R&D experience into a grab-and-go energy chocolate — and why a big-company background changes how you launch. We get into the science of chocolate (conching, flavour optimization, why the same bar tastes different in five countries), the consumer-led product development that got the recipe right in three tries instead of hundreds, and the packaging and branding choices that help the bar stand out in two of the most saturated aisles in the store. Kristyn also shares the deeply personal story that gave the brand its name and brought her back home to Canada. There's a real lesson in here for founders: making something in your kitchen is one thing — knowing the industry, the iterations, and the science behind market fit is what actually moves a brand forward. Find Seven Summit Snacks at https://sevensummitssnacks.com/, on Amazon, and in retailers like Running Room, MEC, and Community Natural Foods. In this episode: From Disney on Ice to food science to global chocolate R&D Inside Cadbury and Godiva: how big chocolate really formulates The origin of Seven Summit Snacks and the story behind the name Consumer-led development: nailing the recipe in three tries Building for two customer types — online buyers vs. in-store shoppers Winning the aisle with packaging, iconography, and clean ingredients A big thank you to CHFA for sponsoring This Commerce Life. CHFA is the voice of Canada's natural health and organic products industry, and their trade shows are where the best emerging brands and buyers connect. Heading to Toronto? Sign up for CHFA East and be part of it: https://www.chfanow.ca/toronto

Women who test positive for bacterial vaginosis (BV) were over three times more likely to also test positive for one or more non-viral sexually transmitted infections (STIs) than women who test negative for BV, according to a new Health Trends® study from Quest Diagnostics and Hologic, Inc., published in the journal Obstetrics and Gynecology Open. Joining me today to discuss the study and what healthcare providers should do based on this information is Beth Marlowe , PhD. Dr Marlowe is Executive Scientific Director and heads R&D for Quest Diagnostics Infectious Diseases and Immunology and co author of the study.   Vaginitis and Sexually Transmitted Infections Coinfections

In this episode, Manna founder Bobby Healy explains how his company designs and operates autonomous electric delivery drones, achieving over 300,000 flights and 97% availability in harsh Irish weather by running operations like a low-cost airline focused on high utilization and efficiency. He discusses why a recent U.S. policy shift and forthcoming Part 108 regulations prompted Manna to go all-in on America, starting with a citywide drone delivery mesh in Tulsa, Oklahoma, chosen for its suburban density, aerospace ecosystem, and pro-drone stance. Bobby outlines why he sees players like Wing, Zipline, and Amazon more as fellow builders in a massive, non–winner-take-all market, and why partnering with aggregators such as DoorDash and Uber Eats is the most efficient go-to-market path. He also covers Manna's recent $50 million fundraise to scale manufacturing, operations, and R&D, and shares his 10-year vision in which drone delivery becomes the dominant, far cheaper, and faster last-mile solution for tens of millions of U.S. suburban homes. Highlights from their conversation include: Introducing Bobby Healy and Manna Overview (0:29) Making Drone Delivery Work Commercially Like a Low-Cost Airline (1:34) Why Now Is the Time for Manna To Enter the U.S. Market (3:28) Why Tulsa, Oklahoma Is Manna's U.S. Launch City (6:19) Competing With Wing, Zipline, Amazon, and Other Drone Players (8:04) Role of DoorDash, Uber Eats, and Aggregators in Manna's Strategy (11:24) Operational Playbook for Launching New Drone Cities (15:23) How Manna Will Use Its Recent $50 Million Fundraise (17:51) Ten-Year Vision for Drone-First Last-Mile Delivery in the U.S. (19:19) Final Thoughts and Takeaways (21:18) Dynamo Ventures is a venture firm backing founders upgrading the physical economy. As intelligence moves into critical infrastructure and technology collides with physics, industry is entering a new era of transformation - the industrial renaissance. Born from the dirt and grit of supply chains and shaped by operations, not spreadsheets, Dynamo focuses on the complex realities of building in the real world. We invest in companies transforming infrastructure, manufacturing, logistics, transportation, and the systems that power global commerce. Dynamo works closely with founders who combine ambition with a bias to action, bringing a builder mindset to venture capital through deep operational insight, systematic pressure-testing and hands-on partnership. Our purpose is simple: to back the relentless shaping the industrial renaissance. Learn more at www.dynamo.vc. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

This week's podcast was recorded live on stage at Latitude Festival 2026, in the Cosmic Shambles Forest of Science and Culture! Helen Czerski is joined by our very own Robert Llewellyn and battery chemist Professor Saiful Islam for a wide-ranging conversation about batteries — covering everything from world-record lemon batteries to five-minute EV charging, sodium-ion tech, battery planes, and vehicle-to-grid power. https://cosmicshambles.com/ 00:00 – Introduction 02:53 – Meet the guests: Robert Llewellyn & Prof. Saiful Islam 03:34 – The world-record lemon battery (and Saiful's quest to reclaim it) 08:54 – What does a battery chemist actually do? 12:42 – Debunking range anxiety: the real stats on UK car journeys 13:34 – From gadgets to vehicles: how battery use has transformed 14:36 – How do you actually measure a "good" battery? 15:21 – Where battery R&D is focused: sustainability, cobalt, and beyond 17:19 – What is a battery? The "electrochemical sandwich" explained 19:38 – BYD's flash charging: 10–70% in 5 minutes 23:31 – Why fast charging is hard — and the safety engineering behind it 28:04 – Battery recycling: Redwood Materials and "second life" batteries 31:54 – Battery passports and tracking materials 34:59 – Sodium-ion batteries: cheaper, longer-lasting home storage 37:34 – Solid-state batteries explained (and why they're the "holy grail") 39:47 – Who's leading the battery race? China's extraordinary scale 42:44 – The tipping point: global crude oil sales are falling 46:34 – Could batteries ever power planes? eVTOL, hybrids & battery gliders 52:57 – Grid storage and vehicle-to-grid: turning EVs into power plants 55:36 – Australia's home battery boom and its impact on coal 57:36 – Audience Q&A Why not come and join us at our next Everything Electric expo: https://everythingelectric.show Support our StopBurningStuff campaign: https://www.patreon.com/STOPBurningStuff Become an Everything Electric Patreon: https://www.patreon.com/fullychargedshow Buy the Fully Charged Guide to Electric Vehicles & Clean Energy : https://buff.ly/2GybGt0 Subscribe for episode alerts and the Everything Electric newsletter: https://fullycharged.show/zap-sign-up/ Visit: https://FullyCharged.Show Find us on X: https://x.com/Everyth1ngElec Follow us on Instagram: https://instagram.com/officialeverythingelectric To partner, exhibit or sponsor at our award-winning expos email: commercial@fullycharged.show EE GREATER LONDON (Twickenham) - 11th & 12th Sept 2026 EE SYDNEY - Sydney Olympic Park - 18th - 20th Sept 2026 #fullychargedshow #everythingelectricshow #homeenergy #cleanenergy #battery #electriccars #electric-vehicles-uk

What does it mean to run product like a mini company, and how do you say no to a multi-million dollar deal when it conflicts with your strategy? In this episode of the CPO Rising series hosted by Products That Count Resident CPO Renee Niemi, ATOSS Software SE CPTO Vikas Seth will be speaking on outcome-driven product management and what the next evolution of the product leadership role looks like. Drawing on seven years growing ID Now from startup to scale-up at 10x revenue, Vikas shares his framework for closing the gaps between customers, go-to-market, and R&D, and why building enthusiasm for the product is the most underrated leadership skill.

Dr. Mohamed is a Technology Executive with over 20 years of experience, uniting advanced R&D and global market execution to architect multi‑million‑dollar Industrial and Generative AI transformations. LinkedIn: http://www.linkedin.com/in/mohamed-salahuddin-innovation CallumConnects Micro-Podcast is your daily dose of wholesome leadership inspiration. Hear from many different leaders in just 5 minutes what hurdles they have faced, how they overcame them, and what their key learning is. Be inspired, subscribe, leave a comment, go and change the world!

Live from Ag in Motion at the John Deere booth, Kristjan and Evan are joined by Deanna Kovar, President of Deere's Ag and Turf Division. She grew up on a Wisconsin dairy farm, started as a summer intern decades ago, and has since worked her way through nearly every corner of the company. It's a chance to talk shop with someone who's both run the numbers on a combine's R&D budget and filled the drill herself. The conversation ends up being less about any single product and more about how Deere thinks. How a company built on iron became just as much a technology company, why that bet on precision ag decades ago still shapes everything they build today, and how they're trying to earn farmers' trust with their data while actually making it useful. Trade policy and tariffs come up; so does the long road from "nobody wanted to touch Operations Centre" to "I can't run my farm without it," and there's real talk about right to repair, retrofitting older equipment, and where the next wave of value on the farm is actually going to come from. It's a wide-ranging, honest conversation from someone who clearly hasn't lost the farm-kid perspective despite the corner office — capped off with a few personal questions that get at how she thinks about leadership, decision-making, and what she'd want for the future of North American agriculture. 

For companies in the defense industrial base, a compliance deadline is not paperwork. It is the difference between winning contracts and watching them stall. In this Brand Feature, Jason LaPointe, Chief Technology Officer at Exostar, and Michael Parisi, Chief Growth Officer at Steel Patriot Partners, walk through what it takes to get FedRAMP ready without cutting corners. Exostar was born out of a consortium that included Boeing and Lockheed Martin, and its FedRAMP-moderate posture lets smaller suppliers keep working on Department of War contracts. How does that work? Instead of moving every server and mailbox into a secure boundary, a supplier inherits roughly 80% of the controls from Exostar, which shrinks the scope of its own CMMC audit considerably. The clock was real. At the time, a November transition date loomed, after which many suppliers could no longer self-attest. That specific timeline has since been paused, but the pressure to prove readiness has not gone away. Exostar needed to show it was FedRAMP-moderate and ready for an audit, and working with Steel Patriot Partners, the team pulled a January target in by nearly three months, not by skipping steps, but by moving with confidence. Why build a new platform instead of retrofitting the old one? Jason LaPointe describes a platform first initiative: build the new compliant home, then migrate customers into it. Trying to modernize inside a live production environment would have been disruptive, so the team built alongside rather than on top, which freed them to re-architect and retool without breaking customers. Michael Parisi frames the engagement as embedding, not staff augmentation. Steel Patriot Partners plugged directly into the product team through daily standups and leadership calls, delivered infrastructure as code and deployment pipelines, and kept the work with US citizens, a requirement once controlled unclassified information is in play. What makes an audit go smoothly? Preparation that extends to how questions get answered. Jason LaPointe compares the audit to a deposition, where an unsolicited comment hands an assessor somewhere new to go. Michael Parisi, who spent years in the assessor's seat and ran the practice for a large C3PAO, explains why knowing the auditors and presenting information cleanly protects the outcome. The business math is unforgiving. Miss the audit window and millions in direct contracts can be exposed, while auditors book out six to eight months. Exostar cleared it with a clean, no POA&M result, and the business is now seeing tailwinds through initiatives like Golden Dome. The lesson Jason LaPointe offers other technology and security leaders is about temperament. Every part of the organization gets touched, from R&D to HR to finance, and the willingness to change quickly becomes the governor on success. Having a clear voice at the table for what good looks like, as Steel Patriot Partners provided, is what accelerates the decisions. This is a Brand Feature. A Brand Feature is a ~30 minute in-depth conversation designed to go deep on a company's story, solutions, and customer success. Learn more: https://www.studioc60.com/creation#feature GUESTS Jason LaPointe, Chief Technology Officer, Exostar Website: https://www.exostar.com/ LinkedIn: https://www.linkedin.com/in/jasonlapointe Michael Parisi, Chief Growth Officer, Steel Patriot Partners Website: https://www.steelpatriotpartners.com/ LinkedIn: https://www.linkedin.com/in/michael-parisi-4009b2261/ RESOURCES Learn more about Exostar: https://www.exostar.com/ Aerospace and Defense solutions from Exostar: https://www.exostar.com/industries/aerospace-defense/ Learn more about Steel Patriot Partners: https://www.steelpatriotpartners.com/ Find Your Path with Steel Patriot Partners: https://steelpatriotpartners.com/find-your-path/ Are you interested in telling your story? ▶︎ Full Length Brand Story: https://www.studioc60.com/content-creation#full ▶︎ Brand Spotlight Story: https://www.studioc60.com/content-creation#spotlight ▶︎ Brand Highlight Story: https://www.studioc60.com/content-creation#highlight KEYWORDS Jason LaPointe, Michael Parisi, Exostar, Steel Patriot Partners, Sean Martin, brand story, brand marketing, marketing podcast, brand feature, FedRAMP, FedRAMP-moderate, CMMC, CMMC 2.0, defense industrial base, DIB, controlled unclassified information, CUI, compliance inheritance, C3PAO, FedRAMP audit, platform modernization, GCC High, Department of War, defense supply chain, cybersecurity compliance Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

Dr. Mohamed is a Technology Executive with over 20 years of experience, uniting advanced R&D and global market execution to architect multi‑million‑dollar Industrial and Generative AI transformations. LinkedIn: http://www.linkedin.com/in/mohamed-salahuddin-innovation CallumConnects Micro-Podcast is your daily dose of wholesome leadership inspiration. Hear from many different leaders in just 5 minutes what hurdles they have faced, how they overcame them, and what their key learning is. Be inspired, subscribe, leave a comment, go and change the world!

IN CLEAR FOCUS: Most new products fail by solving technical problems over consumer needs. Sally Kemkers of consultancy Untapped Innovation reveals how to bridge R&D and marketing. Discussing her new book, "Untapping Innovation," we discover how to drive human-led innovation using The Hero's Journey, uncover hidden habits, and write product stories that align cross-functional teams. Sally also explains how semiotics can help design future-focused products that truly resonate with consumers.  

Dr. Mohamed is a Technology Executive with over 20 years of experience, uniting advanced R&D and global market execution to architect multi‑million‑dollar Industrial and Generative AI transformations. LinkedIn: http://www.linkedin.com/in/mohamed-salahuddin-innovation CallumConnects Micro-Podcast is your daily dose of wholesome leadership inspiration. Hear from many different leaders in just 5 minutes what hurdles they have faced, how they overcame them, and what their key learning is. Be inspired, subscribe, leave a comment, go and change the world!

Phil Chang and Kenny Vannucci sit down with Sylvia Bennett, founder of SURCH Foods (that's S-U-R-C-H), a B2B search platform built to solve a problem the food industry has quietly lived with forever: there is no good way to find ingredients, packaging, co-manufacturers, or expertise.  Sylvia spent her career in the industry — starting in a Sobeys bakery, then product development and private label at Sobeys and Walmart, then R&D at a cookie plant — and kept watching sourcing managers dig through old business cards because "I know a guy" was the only system available. We flipped this episode on its head. Instead of starting with Sylvia's story, we start with the platform, walk through it live, and then back into why someone with her background decided to build it.   The honest takeaway: a search engine for food ingredients shouldn't be revolutionary in 2026. But it is — and it only works if the industry actually shows up and gets listed. Find Sylvia at surchfoods.com or sylvia@surchfoods.com.

Congress Stock Ban Games, NDAA Section 219, and the Episode's Running Bits Dan and Corey focus on Congress and political theater, digging into the House-passed ban on lawmakers (and families) trading individual stocks, praising the idea but warning about loopholes, keeping existing holdings, and exempting the president and vice president; they highlight Thomas Massie's point that Republicans attached a voter ID provision to force Democrats to vote no and then run midterm ads claiming Democrats opposed banning insider trading. They also break down NDAA Section 219, arguing it makes expanded U.S.–Israel technology, R&D, and defense cooperation into law, criticizing Ben Shapiro's framing and debating whether “merging parts of the military” is overstated but still a sovereignty concern. The episode ends with heavier humor and clips, including a New York candidate railing at Ben & Jerry's, a Mark Levin–Lindsey Graham tribute, Pride corporatization commentary, virtue-signaling satire, and a viral treehouse-vs-permits story, plus assorted banter and show drops. 00:00 Welcome Back Libservative 00:56 Trump Speech Hangover 02:52 No Prosecutions Theater 05:08 China Data Nothingburger 07:51 Norway Oil Side Quest 12:59 Cop Dream Detour 16:55 NDAA Section 219 23:18 Shapiro Doublespeak 31:44 Congress Stock Ban 36:11 Voter ID Poison Pill 39:20 Parties Both Right Wing 40:31 Dating Politics Banter 41:45 Voter ID Debate 43:21 Camera Blur Banter 44:13 Tyler Robinson Hearing 46:54 Blurry CCTV Narrative 52:51 Dusty Barrel Questions 56:30 Lance Twiggs Testimony 01:05:54 Reasonable Doubt Talk 01:09:13 War Narrative Returns 01:09:27 Midwit Talking Heads 01:11:52 Russia China Blame Game 01:18:25 Michigan Primary Preview 01:19:48 Haley Stevens Cringe Clip 01:23:27 Senate Ads Breakdown 01:26:00 PAC Money and FARA 01:27:00 Blakeman Ben and Jerrys Clip 01:30:42 Levin on Lindsey Graham 01:36:10 Pride Gets Corporatized 01:40:45 Activism as Identity Trap 01:42:45 Virtue Signaling Skit 01:48:39 Politics as Sports Culture 01:52:55 Treehouse vs Red Tape 01:58:29 Neighbor Fence Line Rant 02:01:25 Closing and Plugs  

"Everything starts with trust." "Start with listening." "Culture is nothing different from the collective behaviour of people." "Dare to do things differently." "Leadership is giving direction and inspiration to bring out the best possible for the organisation." Brief Bio Karl Deppen is President and Chief Executive Officer of ARCHION Corporation, the new holding company bringing together Mitsubishi Fuso Truck and Bus Corporation and Hino Motors. ARCHION began operations on 1 April 2026, combining two major Japanese commercial-vehicle brands to strengthen scale, competitiveness and investment in decarbonised transport, autonomous driving and digital mobility solutions. Deppen began his career more than 35 years ago with the former Daimler-Benz Group in procurement and supply-chain management. His international assignments have included Portland, Istanbul, Beijing, Stuttgart, Brazil and Japan. He has held roles spanning procurement, logistics, product management, leadership development, finance and general management, including Regional CFO for Daimler's China operations, responsibility for finance across Mercedes-Benz assembly plants and R&D operations, leadership of the South American truck business, membership of the Daimler Truck Board of Management with responsibility for Asia, and President and CEO of Mitsubishi Fuso. His first Japan assignment began in 2002, when he became one of the earliest Daimler employees to join Mitsubishi Fuso after Daimler acquired a stake. That experience, followed by a return to Japan two decades later, shaped his conviction that adaptability in Japan begins with patient listening, understanding context and resisting the urge to impose pre-formed answers. Narrative Summary Karl Deppen's leadership journey is unusually broad, even by the standards of a global automotive executive. Over more than three decades, he has moved across functions, continents and cultures, building experience in procurement, supply chains, product management, human resources, finance and operational leadership. That range now matters enormously. As President and CEO of ARCHION Corporation, he is leading the integration of Mitsubishi Fuso and Hino at a time when the commercial-vehicle industry is being reshaped by decarbonisation, autonomous driving, digitalisation and intensifying Asian competition. His first encounter with Japan came in 2002. He arrived without Japanese language ability or previous exposure to the country and entered a highly operational environment at Mitsubishi Fuso. The language barrier sharpened his ability to observe, draw, use numbers and read what people were trying to communicate. More importantly, it taught him patience. Trust was not created through positional authority or a perfectly delivered executive message. It emerged through repeated interaction, careful listening and a visible effort to understand. That lesson became central to his approach when he returned to Japan roughly twenty years later. Rather than assuming that earlier experience gave him a complete map, he deliberately treated the country and company as changed environments. Japan is a high-context culture, and effective leadership requires understanding the operating system beneath visible processes. What can initially look slow, complicated or overly cautious may reflect customer requirements, stakeholder commitments, quality expectations or prior experience. Deppen's practical response is to ask why repeatedly, suspend judgement and speak directly with employees, customers, suppliers, banks and other stakeholders before changing the system. He also sees genuine strength in Japanese decision-making. Consensus building through nemawashi and formal approval mechanisms such as the ringi-sho can extend the front end of a decision, but once the team is aligned, execution can be exceptionally fast. The leadership opportunity is therefore not to discard consensus, but to shorten decision cycles while preserving commitment and execution discipline. This is decision intelligence in practice: understanding not only what decision should be made, but how the organisation must be engaged so that the decision becomes real. Trust, transparency, empowerment and accountability form the backbone of Deppen's leadership philosophy. Employees need enough information to understand direction, freedom to act within clear expectations and accountability for outcomes. Difficult moments are decisive proof points. Leaders either build trust by listening to bad news and solving problems, or destroy it by shooting the messenger. In a zero-defect culture, fear can drive mistakes underground. Psychological safety is therefore not softness; it is an operating requirement that allows risks, missed targets and errors to surface early enough to be corrected. Innovation follows the same logic. Deppen supports structured improvement circles, skip-level dialogue, pilots and repeated lessons-learned reviews. The goal is not reckless experimentation, but controlled learning: try something on a limited scale, assess what worked, identify root causes when it did not, refine the process and then scale. For artificial intelligence, he argues that simply digitising an existing paper workflow rarely changes the needle. Companies need end-to-end redesign at the data layer, potentially supported by tools such as digital twins, automation and integrated decision systems. Ultimately, Deppen defines culture as collective behaviour. Purpose, vision and values matter only when translated into visible practices: what people do, what they refuse to tolerate, how they respond to mistakes and how they treat one another. His advice to leaders arriving in Japan is therefore disciplined and practical: start with listening, understand the context, assume there is a reason behind existing behaviour, learn enough language to widen access to the culture, and build the trust that makes change possible. Q&A Summary What makes leadership in Japan unique? Japan combines a high-context communication environment with strong expectations around quality, detail and collective commitment. Decisions are rarely just agreements between two senior people. They are embedded in a wider network of stakeholders, informal nemawashi and formal mechanisms such as the ringi-sho. This can make the decision phase appear slow, particularly to executives from low-context cultures. The advantage is that once consensus is genuine, implementation can move with impressive speed because the people responsible for execution understand the rationale and are already committed. Effective leadership therefore respects the social architecture of the decision while working to reduce unnecessary delay. Why do global executives struggle? Global executives often arrive with a strong track record and an assumption that proven headquarters practices should be transferred quickly. The visible process may look inefficient, bureaucratic or resistant to change. Deppen cautions that there is usually a good reason for why people do things the way they do. The reason may sit with a Japanese customer, a regulatory expectation, a quality requirement or an unspoken stakeholder agreement. Leaders struggle when they judge before they investigate. Direct engagement with employees, customers, suppliers, banks and partners helps reveal the operating context and prevents premature restructuring based on incomplete information. Is Japan truly risk-averse? Japan often displays high uncertainty avoidance, especially where mistakes can damage reputation, customer trust or career security. However, Deppen distinguishes caution from an unwillingness to improve. The greater danger is a zero-error aspiration that encourages people to conceal problems. Leaders must create psychological safety for employees to raise a flag, admit that a target may be missed and request help. Innovation becomes possible when experiments are bounded through pilots, resources and review points. A failed pilot should produce a structured lesson, not a search for someone to blame. What leadership style actually works? Deppen's approach begins with active listening and develops through communication, transparency, empowerment and accountability. He argues that people quickly sense whether a leader is genuinely trying to understand them. Trust grows through meaningful dialogue and is tested when conditions become difficult. Leaders must clarify expectations, provide freedom to operate and hold people accountable without turning accountability into punishment. He summarises his model as TEAM: trust, empowerment, accountability and mutual ambition. Shared ambition matters because a large, team-dependent organisation cannot succeed if the CEO's targets are not believed or owned by the wider workforce. How can technology help? Technology can accelerate transformation, but only when leaders redesign work rather than automate yesterday's process. Applying artificial intelligence to an existing paper flow may produce marginal efficiency without changing the business outcome. A stronger approach examines the process end to end, starts with the data layer and asks how the workflow would be designed if digital capability were native from the beginning. AI, automation, decision intelligence and digital twins can improve speed, visibility and scenario testing, but the human organisation still needs the trust and learning discipline required to adopt them. Does language proficiency matter? Japanese proficiency helps because language opens a window into culture, relationships and assumptions that may not be stated directly. Deppen also notes that Japan contains substantial English capability, although employees may hesitate to speak when they fear exposure or embarrassment. Leaders can lower that barrier through patience, written follow-up, visual explanation and an environment in which imperfect language is accepted. Language should improve communication, not become an excuse to avoid it. What's the ultimate leadership lesson? The central lesson is to start with listening. A leader entering Japan should temporarily set aside fixed ideas about what is right or wrong and first understand the existing operating system. Culture is the collective behaviour of people, so change depends on the practices that leaders repeatedly model and reinforce. Purpose and values provide direction, but trust turns them into action. Leadership succeeds when people understand the direction, feel empowered to contribute, accept accountability and share the ambition to achieve something together. Author Credentials Dr. Greg Story, Ph.D. in Japanese Decision-Making, is President of Dale Carnegie Tokyo Training and Adjunct Professor at Griffith University. He is a two-time winner of the Dale Carnegie "One Carnegie Award" (2018, 2021) and recipient of the Griffith University Business School Outstanding Alumnus Award (2012). As a Dale Carnegie Master Trainer, Greg is certified to deliver globally across all leadership, communication, sales, and presentation programs, including Leadership Training for Results. He has written several books, including three best-sellers — Japan Business Mastery, Japan Sales Mastery, and Japan Presentations Mastery — along with Japan Leadership Mastery and How to Stop Wasting Money on Training. His works have also been translated into Japanese, including Za Eigyō (ザ営業), Purezen no Tatsujin (プレゼンの達人), Torēningu de Okane o Muda ni Suru no wa Yamemashō (トレーニングでお金を無駄にするのはやめましょう), and Gendaiban "Hito o Ugokasu" Rīdā (現代版「人を動かす」リーダー). In addition to his books, Greg publishes daily blogs on LinkedIn, Facebook, and Twitter, offering practical insights on leadership, communication, and Japanese business culture. He is also the host of six weekly podcasts, including The Leadership Japan Series, The Sales Japan Series, The Presentations Japan Series, Japan Business Mastery, and Japan's Top Business Interviews. On YouTube, he produces three weekly shows — The Cutting Edge Japan Business Show, Japan Business Mastery, and Japan's Top Business Interviews — which have become leading resources for executives seeking strategies for success in Japan.

In recent months, the open vs closed, and US vs China discussions on model ownership and sovereign/local AI have heated up to a fever pitch. So it is very very good news that Poolside AI are finally emerging with new models, like Laguna S 2.1, that are beating Thinking Machines' recent release nearly 10 times their size.Poolside's recent tech report got a lot of praise due to their level of detail, and Vibhu first covered Laguna's recent technical report on our paper club:From spending $12 million building language models for code before the world cared to creating a Model Factory that can take a model from pre-training to release in eight weeks, Eiso Kant has spent more than a decade betting that code is the path to AGI. In this episode, the Poolside co-founder joins swyx and Vibhu to explain why ChatGPT felt like vindication, why Poolside embraced open weights and open research, and why he would rather live in a world with 100 foundation model companies than five even if Poolside were one of the five.We go deep on Poolside's Model Factory: the engineering systems behind 10,000–20,000 experiments per month, streaming data directly into training, reproducible experimentation, low-precision compute, and agents that increasingly write code, launch jobs, evaluate results, and modify the pipelines used to train future models. Eiso also unpacks their recent launch Laguna S, why persistence, verification, and backtracking may matter more than raw intelligence, how much capability remains inside smaller models, why reinforcement learning will move earlier into pre-training, and why next-token prediction is still extracting too little from the web.We also discuss model-harness co-design, Poolside's path from coding agents to AGI, why Eiso thinks MCP and traditional tool calls are “stupid,” the real economics behind frontier-model training, Poolside's $500 million raise, open-source AI, regulation, NVIDIA and TSMC's influence, engineering productivity in the agent era, high-agency teams, and hiring at Poolside.We discuss:* How Andrej Karpathy's RNN work inspired Eiso to start building language models for code in 2015* Why Eiso spent four years and $12 million pursuing an idea before the market cared* Why ChatGPT felt like vindication and brought Poolside back to open source* Why Eiso would prefer 100 foundation model companies over an oligopoly of five* The difference between releasing open weights and publishing genuinely open research* Why Poolside deliberately built a global research organization outside the Bay Area talent war* Why model building is ultimately 90% engineering* The Model Factory: Poolside's end-to-end system for rapidly training and improving models* How fewer than 70 researchers run roughly 10,000–20,000 experiments each month* How Poolside moved from six-month model cycles to five- and eight-week launches* Why streaming data directly into training unlocked faster experimentation* How immutable data, versioned code, and reproducibility enable rigorous model research* Why Eiso wants capable researchers to leave their labs and become Poolside's competitors* Why 95% of model building can be reduced to better data or compute efficiency* Laguna S and why persistence, verification, and backtracking can outperform raw intelligence* Why smaller models may handle far more knowledge work than previously expected* Why reinforcement learning will move earlier into pre-training* Why next-token prediction is still failing to extract enough knowledge from the web* Why distillation and environments have become the AI industry's favorite “drugs”* Why mid-training is really an early form of curriculum design* Low-precision training, networking bottlenecks, and the next gains in compute efficiency* Laguna S: 118 billion total parameters, 8 billion active, and eight weeks from training to launch* Why model builders can often evaluate a new checkpoint within its first 30 minutes* Model versus harness: where agent capabilities actually come from* Why Poolside sees coding and long-horizon software tasks as a path to AGI* Why Eiso thinks MCP and traditional tool calls are “stupid”* Why future agents will write scripts instead of choosing from dozens of predefined tools* The case for minimal harnesses, containers, and model freedom* Why Poolside is prioritizing vision but does not expect to work on audio soon* Why language may be the most compute-efficient modality for encoding knowledge and reasoning* The real cost of model development and why the final training run is anticlimactic* The story behind the Poolside name and why it represents refusing to lower ambitions* How Poolside raised $500 million while investors still questioned whether AGI was real* Why intelligence could become the world's most demanded and commoditized resource* When open models may become too capable to release without restrictions* Why unilateral AI safety does not work in a globally competitive environment* How regulation could accidentally lock in an oligopoly of two or three AI companies* NVIDIA, TSMC, and the hardware systems underpinning foundation-model progress* Why reinforcement-learning wall-clock time is one of Poolside's biggest bottlenecks* Why Poolside trains models from scratch instead of simply distilling larger models* How AI changes the way companies should measure engineering productivity* Why agency may become the most important quality for employees in the AI era* How leaders align high-agency people through shared goals and clear constraints* Hiring across research, post-training, pre-training, architecture, evals, and engineering at PoolsideEiso KantLinkedIn: https://www.linkedin.com/in/eisokantX: https://x.com/eisokantPoolside: https://poolside.aiTimestamps00:00:00 Introduction00:00:54 Karpathy, RNNs, and Building Code Models Before Transformers00:02:26 The $12M Failure and ChatGPT Vindication00:03:39 Open Source and the Case for 100 Foundation Model Companies00:09:22 Open Weights, Open Research, and Poolside's Global Team00:16:04 The Model Factory: Why Model Building Is 90% Engineering00:20:19 Agents, Automated Experiments, and Early Signs of RSI00:24:04 Streaming Data, Reproducibility, and Scientific Rigor00:30:35 Creating More Foundation Model Companies00:36:07 Laguna S: Persistence vs. Raw Intelligence00:43:01 Reinventing Pre-Training, RL, and Curriculum Design00:52:33 Low-Precision Training and Squeezing More From Smaller Models00:58:37 Model Harnesses, Coding Agents, and the Path to AGI01:09:26 Why MCP and Traditional Tool Calls Are “Stupid”01:13:04 Vision, Multimodality, and Why Language Still Matters01:18:15 Scaling Models and the Real Economics of Training01:20:40 Why Poolside Is Called Poolside and Raising $500M01:27:37 Open Models, AI Safety, and the Risk of an Oligopoly01:33:53 NVIDIA, TSMC, and the Reinforcement-Learning Bottleneck01:41:52 Smaller Models, Distillation, Engineering Productivity, and HiringTranscriptIntroduction: Eiso Kant, Poolside, and Open ModelsSwyx [00:00:00]: All right, we're here in the studio with Eiso Kant from Poolside, together with Vibhu. Welcome.Eiso Kant [00:00:08]: Thanks. Thanks for having me, guys. Good to be here.Swyx [00:00:10]: Yeah, fresh on the plane. You texted me, you were like, “Hey, I'm on my way to SF.” I was like, “You're on a plane right now, right?” Like, hey.Eiso Kant [00:00:16]: I know. After I texted you, I realized that probably coming in with major jet lag was gonna offer some fun experiences today, but let's do it.Swyx [00:00:23]: I mean, I think the thing I would tell guests is that they don't have to prepare that much because if you're truly working on this every single day, then even, like, what you hazily remember is going to be new for a lot of the audience that don't live in your world every day, right? so 10 years ago, you did a talk at Google Slush, talking about the democratization of AI. and, now here you are, like, open sourcing an incredible new model that we're gonna talk about. But I guess, like, what got you into democratization of AI? Like, it's not obvious from your LinkedIn or something.From Karpathy's RNN Post to SourcedEiso Kant [00:00:57]: No, it's not at all. I don't think it's obvious how I got in this space. I owe getting into this space to Andrej Karpathy.Eiso Kant [00:01:05]: In 2015, he wrote an article called “The Unreasonable Effectiveness of Recurrent Neural Nets.”Swyx [00:01:10]: Neural Nets, yep.Eiso Kant [00:01:11]: And that article, I read it, and I pivoted my startup at the time overnight to working on RNNs, and later LSTMs and Transformer models to be able to write code. If you go to this article and you scroll down, you can start seeing, like, this was the precursor to what ended up becoming language models. So, at least when he was character-level language models that were starting to predict letters, he has an example out here. There's a little Paul Graham generator, and you can read it, and the text makes sense, but it doesn't. and there's a little-- There's an example of code a little bit further down. Yeah, so Shakespeare.Swyx [00:01:47]: Shakespeare.Swyx [00:01:49]: CoolEiso Kant [00:01:49]: And for some reason, I read this, and I went down the rabbit hole of learning everything I could about RNNs and LSTMs, right? This is Transformer paper. And I had built a completely unreasonable belief, that neural nets should be able to generalize to anything and everything, and that language should be able to generalize, to a lot of things that are intelligent and the ability to write code. And so I started building Sourced, which was a fully open source company trying to build, what we used to call machine learning on code, language models on code. And we spent about four or five years on this, till the end of 2019. And that sounds really cool today, but back then, no one cared.Eiso Kant [00:02:29]: Right? Like, no one cared. We were in the dark. Like, we did things along the way. We tried applying convolutional neural nets to, like, the structure of code. We were. when attention came out, we were applying it to LSTMs, and then the Transformer paper came out. And it - it wasn't obvious, and what we missed throughout that entire journey, that we were on the right track, but we should have just kept scaling up. And today, to all of us, the scaling laws and scaling up seems like the most obvious thing. But having spent four or five years of my life on working on language models on code, it wasn't obvious. So I have a lot of respect to folks at Google and OpenAI and others who took that confidence and kept going. we failed ultimately at the time, and it was, like, biggest failure of my career, right? You blew $12 million of investors' money, which was a lot back then.Swyx [00:03:18]: Yep.Eiso Kant [00:03:19]: You spent, still a lot, but, And you spent years with, like, a group of 40 people just obsessing over this problem. And life took a different turn, And it was, and family became a focus, and I kept my heads down and really, didn't really look at language models for the following two years. big mistake considering Following years are gonna be really interesting. And then ChatGPT came out And it was like a vindication. It's like people started texting me. I found, like, my old, work decks and these old talks. And throughout that whole journey, we,ChatGPT, Vindication, and Returning to Open SourceEiso Kant [00:03:56]: We really had a strong point of view at the time that, like, as you're building more capable intelligence, it should be open and open source.Eiso Kant [00:04:04]: When we started Poolside, that wasn't the case at all, and I wanna be very open about it. When we started Poolside, we were like, there was a premise of two things. One is this technology is not gonna stop compounding in capabilities. I think to most people obvious today, but three-plus years ago when we started, most people were still arguing if these were stochastic parrots or not.Eiso Kant [00:04:23]: And the second was that reinforcement learning was gonna be the biggest driver for LLM capabilities. Today, very obvious. Three years ago, was not an opinion held or direction held at either OpenAI or Google or Anthropic or others. And so people looked down on us a little bit. They were like, “ is this really gonna work?” And so we just started working the problem, and we never really thought about open source again. We just kept our heads down and we built our, like, knowledge, understanding from scratch, right? We didn't roll out of an existing lab. So we picked up the papers and started writing code and figuring things out.Eiso Kant [00:04:59]: And it wasn't until the beginning of this year that me and my founder, Jason, picked up the open source conversation again.Eiso Kant [00:05:07]: And if you go back to some of the early things on our website, it was very straightforward. It was we wanna get to AGI, we wanna support a world of abundance, and we wanna be the first company that gets there.Eiso Kant [00:05:20]: But we started talking at the beginning of this year because it became obvious that the world was going in a direction that was starting to like, pick at us a little bit. Like, it didn't, this didn't happen overnight. It was, like, a little bit we were seeing this and we're like, “Okay, The world's going down a path.” And Throughout this journey, there was something that I used as a, as an analogy or thing. So I said well, if I go back to back in those days, 2015 or 2016, we're working on this, and I picked up a fi book off the shelf, and I was reading the book about 2035. AGI is achieved, and the story would be over the following, decades. And it would have that first chapter where everyone's trying to figure things out. You'd get the chapter of ChatGPT coming out And then you would get to the chapter where the world was at a fork in the road, and the one that it picked was one where three or four or a handful of companies were going to create all of intelligence moving forward.Eiso Kant [00:06:21]: And when I thought about that story, it felt like a dystopian fi book, not a utopian fi book. And the reality is, I'm a utopian fi guy. Like, and so We took a step back and said, “Hey, can we play a role here?” Now it was easy for us to do so because we were not at the frontier.Eiso Kant [00:06:41]: If we were at the frontier, I don't think we could have changed our mind. and I don't mean this like it's when the moment there's too much capital involved, too much expectations, you've built up things, right? We're a small team, just improving and improving. And so we knew that we could make that decision now, but it would be a lot harder to make as we got closer and closer to the frontier and caught up to others. And did a lot of soul-searching and a lot of conversations, and said, “No, this makes sense,” Even if there's big unanswered questions, like how the hell do you build a business model with foundation models about open source? Big open-ended question that we do not fully have the answer to yet, right? At what point do you no longer wanna release open source models because misuse of models has, real potential risks associated with it? how is the government gonna respond to open source? but I think it all just came down to one thing, and I'll stop the monologue, is the fact that I rather live in a world that has 100 foundation model companies than a world that has five, even if I was one of the five. And the smallest and most meaningful contribution we can make for 100 to exist is to open up our research and open up, like, our weights right now and figure out along the way how we can, like, do more.Neo-Labs, Model Choice, and the Token EconomySwyx [00:08:01]: Yeah. I think if anything, over the past three years, that has become a bit more true. you are one of a cohort of Neo labsEiso Kant [00:08:10]: YeahSwyx [00:08:10]: That people are now calling that. And, we're, we're doing this on the day that Thinky launched their, new model and you are outperforming them on their, on some benchmarks that they released, right? Like, they just don't have it yet. so it goes to show that I think, like, this is one of those things where, like, there is room for multiple players, and you are seeing a little bit more of the future. Maybe more like 20, not 100, but, like, you are one of the 20.Eiso Kant [00:08:36]: I really hope so, right? I think we I'm, I'm excited about their release, and I'm excited about everyone releasing because, like, ultimately, like, choice competition is both gonna drive progress in the right direction. But the fact that like, we create models and while we all, drink out of the same well of data effectively, we do introduce very different behaviors and biases in our models. Some are intended biases, some are completely unintended biases.Swyx [00:09:03]: Yeah.Eiso Kant [00:09:03]: And if we shape up in an ecosystem in the world where open models are gonna be a part of the token economy, like, I don't think there's any question about it anymore Then we want to be able to live in a world where companies, countries, people can choose and say, “Hey, I am most aligned and I trust most this provider for these things.”Swyx [00:09:25]: Yeah.Vibhu [00:09:26]: I think more than just one of the 20 Neo labs, up until recently, most of open source innovation was coming from the Chinese labs, right? So there's the DeepSeek of the West. Is it today? Okay, maybe it's thinking machines reflection, but there aren't many, right? So, one of the things you guys started in France, Europe, but very much now you're taking that American standpoint and more than just that, the point is the Chinese models that we see, they're not super open research. the work you put out is, I think, some of the best. So every few months you get not only frontier models, but also here's a breakdown blog, paper, technical report of here's everything for state of the art to build, frontier intelligence and you're filling that gap too, right? So not just only open weight, not just Western, but also pretty open research.Open Weights vs. Open ResearchEiso Kant [00:10:20]: No, I appreciate it. Look, I think it's, I think it's the most meaningful contribution, right? Weights are a binary. Let's call them what they are. Yes, we can modify them, we can change them, but, like, giving someone the weights does not allow them ultimately to recreate what you're doing, right? And so now there's challenges around releasing data sets, challenges around like releasing certain things, but being able to share your research, like, right, how do we do it? What are the lessons we learned that we spent, tens of thousands of experiments of compute on? I think very much so. One correction though, Vibhu, and I say this because it's been haunting us for quite a few years. We from day zero were an American company.Swyx [00:10:55]: Yeah. They movedPoolside's Global Team and American Company StorySwyx [00:10:56]: To France.Eiso Kant [00:10:56]: So the story once and for all is very. We start as an American company. We have always been an American company, and early on we made a very conscious decision. We said, “We're not gonna hire any researchers in the Bay Area. We're gonna look for talent everywhere else in the world.” and that is everything from Middle Americas, Seattle to, Serbia, and to Taiwan and Singapore and other places. And it was because we took a view that this was gonna become a talent war for this, and I think it has over the years now. Three years ago, that wasn't fully obvious yet. I think today it very much is. And we also realized that, like, some of the world's most capable people with, like, the most interesting, innovative ideas were not just gonna be here. And so it led us to create like a fully remote company. and we ended up opening an office in Paris and London and different places and we have a lot of the team in the US and a lot of team outside. But we always took this view of like, we're an American company, but if we want the best of the best to work with us, we need to take a global view. Now we do also have people here in Silicon Valley, like the company's grown and others, but I think one of the things that, it slowed us down at the beginning, but it has sped us up now, and it's why you're seeing like the progress, I think, on our models and the cadence at which we release, is because we didn't roll out of an existing lab. Right? we didn't, we didn't have a lot of the information that's freely flowing around here at the time. We just took this point of view as like, “Okay, well, let's just work the problem. Let's just go and, like, read the few papers that are out there, and let's just figure this stuff out.” And we made some hilarious mistakes in model training because of that over the yearsEiso Kant [00:12:35]: Like especially in the first 12 months. there's a few that I think still haunt me and scare me. We can talk about them later. but it created a, like, a resiliency and persistency in the team, right? with extremely few people have left us over the years, that, like, told us, “Okay, we can do this.” When we first wrote our first training code base completely from scratch, it wasn't a fork of any open source. It was just like, “Okay, let's build it from scratch.” I remember we had this one moment where we spent three weeks working out an optimizer bug. Like, it was like training just couldn't get stable. We, like, obsessed over it, and we thought, like, maybe we were wrong. Maybe we should have just forked this repo, or we should have. But then when we solved it, I still remember at the time we were like five people in the company. when we solved it, we were like, “Oh, we can do things,” like if we're just willing to work hard. and I think that culture with a very strong engineering bias has helped us, like, get to where we were. And so there's this notion of open source and talent and these things. I think we, We just took different decisions from a different starting point. and I think we are lucky. I do want to definitely call it lucky. And there was a lot of hard work at the team that now, like, that's starting to show up in results.Swyx [00:13:52]: Just ‘cause we probably won't revisit this again, but, and this is a fun recruiting challenge if someone knows the answer. What was the bug? And then we won't tell the solution, but we'An Optimizer Bug and the Value of Building From ScratchEiso Kant [00:14:01]: So the - This - You're gonna test my memory here,Swyx [00:14:04]: Oh, okayEiso Kant [00:14:04]: So but I thinkSwyx [00:14:05]: DirectlyEiso Kant [00:14:05]: I think I can recall. So if you, so if you look at, So if you take like Adam as an optimizer, you have epsilonSwyx [00:14:12]: YeahEiso Kant [00:14:13]: Which is, right, like in the denominatorSwyx [00:14:14]: Momentum and weights. YeahEiso Kant [00:14:15]: Is exactly, in the denominator. And at the time, if I recall, you looked at like the early Llama papers and things like that. People were juicing epsilon, like, quite a bit. Like, they were, like, adding, I don't know if it was E minus four or whatever, like a high value for epsilon.Eiso Kant [00:14:31]: And if you think about this during training, it's like a bit weird and counterintuitive that we're adding noise to our optimizer by just adding effectively, like, a random number in the denominator, right? Like behind the decimal point. And I don't recall the exact bug, but it had - What I remember is once we solved it, we no longer had to juice epsilon as much as, like, was happening in the Llama paper and other places. and it was like one of those fundamental moments where we had trusted this paper that was out there, and we're like, “Oh, no, it has to be this way. It has to have this high value of epsilon.” But it made no sense to us intuitively. Like, why do you have to have this so high? Like, if you're just trying to avoid division by zero, why can't the value be extremely small? and that was like one of those moments where you realize like, okay, finding things out from scratch yourself builds a better intuition. Because the one thing you learn very quickly with model building is that your intuitions that you start with are gonna get beaten up so hard.Eiso Kant [00:15:33]: Right? Like - It's such an experimental science, that the things that seem obvious, you very quickly get to learn, like, you were wrong, and hopefully you figure out why, and sometimes you don't even.Swyx [00:15:45]: Yeah. yeah, so, one of the reasons that you, when you released your new models, Vibhu got really excited. I mean, everyone got really excited. But Vibhu led our paper club on it, and you guys sawEiso Kant [00:15:58]: YeahSwyx [00:15:58]: Obviously. maybe talk through some lessons learned in that, whatever you can disclose. we can focus on the model factory stuff, whatever you think is a good starting point.Model Building as EngineeringEiso Kant [00:16:08]: So I would say that our view from very early on in the company was that model building is ultimately 90% engineering.Eiso Kant [00:16:18]: And I think we all know it in the industry because if you look at where's every researcher spending their time, they're spending their time writing code, right? Looking at data and writing code. And so we said, okay, The state at the moment, like three years ago, was bash scripts and Slurm and spaghetti code bases for training and, like, data pipelines that were patched together. And we looked at this and said, “Well, ultimately, model building is a process.” You're going from raw data, right? Like training raw material, the web, et cetera. you're doing a whole bunch of filtering, cleaning up, transformations, analyzing. These days, that's, far more complex than it was three years ago. then you're training a model, which is effectively a large distributed systems problem, right? Across hardware that has still-- It's become a lot more reliable. It was extremely flaky back then. and now with every new generation, we get our new sets of challenges. And then you go into the next stages, right? There was no training back then, but, like, you got, your post-training and then your reinforcement learning. And so we looked at this and we said, “Well, this looks like an industrialized process. This looks like an end process, that every single part of it has its machinery,” right? If it's your big data pipelines, if it's your crawling ingestion of the web, if it's your, large-scale distributed training, and then you've got your reliability. And we said, “Well, why don't we take some of the world's smartest distributed systems engineers that we knew and make them part of the process of research from day zero?” Not retrofitting it later on, but, like, really from the beginning. And that became our model factory. And so our model factory started with a handful of components. Today, it's thousands of components, and I try to equate it to, if you think about, like, someone who was at the very early days of Foxconn, if they had been there for the following, decade, they would be able to rebuild Foxconn because they saw every decision that led to building that system and all the complexity. If you and I walk into Foxconn today, no chance.The Model Factory and Experiment VelocityEiso Kant [00:18:18]: Right? Because we don't have the lineage and history of decisions that led to that. And so we built early on from the beginning- with a team that really understood that, well, the metric that we are optimizing for is the speed of an idea from a researcher to an experimental result that we can trust to then being part of the next model training.Eiso Kant [00:18:42]: And in the. And because it's such an experimental science, ultimately, in the beginning when it wasn't that complex, you could patch your way around it, right? But now, at any foundation model company, you are running. I mean, we're a small team, right? We're less than 70 researchers, another 35 engineers. and we are running, I haven't checked the latest count, but far more than 10,000, maybe 10 to 20,000 experiments a month that we cut. And so if you look at that scale of every model run that is, like it's ultimately it's, it's you need to be able to trust it as an infra problem. And so what we have now done over the years is gotten really good at that, and just by working it and improving it and obsessing over those end decisions. So now what that means is that you looked up Laguna XS 2 that we launched. It was five weeks from the beginning of training to launch. The model that we're gonna talk about today was eight weeks from start of training, to launch. We started the next model literally yesterday because we now finished the post-training required for the model we're launching, next week or by the time this comes out today. and we move that compute to the much larger Laguna M model that we're now training. And so the model should be an artifact of someone's process. It shouldn't be really a thing in itself. Like, and we treat this like the way you would look at like a SpaceX factory where, yes, the first rocket, really hard to build, but the much harder challenge was building the factory. And now they're rolling off, and no one is really thinking about the next launch anymore. So it's just another launch, it's another launch, another rocket comes off. And that's what we're trying to do with model building.Eiso Kant [00:20:22]: And what has been, which was not planned from day zero, it was in the back of our mind like this will happen one day, is that when you build a really good end model factory with really good APIs and really good engineering systems, Well, what is it perfect for? It's perfect for agents.Agents Inside the Model FactoryEiso Kant [00:20:40]: Because agents are now starting to take over more and more work in our model factory.Vibhu [00:20:43]: Yeah.Eiso Kant [00:20:44]: So I look at the screens when I walk, like when we're, we come together, in our monthly, we do monthly onsites, and I walk behind people's screens and I stop by and I talk to our researchers. And the default is all of these different agents running on their screen that are writing the code. They're launching the jobs. They're evaluating the results that are coming back from the model runs. They are, making the changes. And we're still in the driver's seat. We're still coming up with the ideas. We're still helping with the debugging. But more and more, and this is right now very profound on the data side of our pipelines in both pre and post and the synthetic data pipelines, it's starting to become more on the architecture side as well. You're starting to see these twinklings of what RSI is gonna look like.Eiso Kant [00:21:27]: And that's. So when we talk about, like to your question about our models, every talk about the model factory, And my coolest example of these things is always that when we kick off a new run, doesn't matter if it's a training like big run or if it's now a post, like one of 10 post-training versions we do for like release or many experiments, is that at any given moment, the changes that somebody made that they had experimental results from the day before make it into that run.Eiso Kant [00:21:57]: So there's not like a cutoff 90 days before. Like no, it's like literally from that moment because we can now trust the machine enough. And then you also have to invest in the reliability. So one of my favorite metrics about like Laguna S is that there was no call events, Right? Like completely zero. And we haven't had a meaningful call event, like something to wake up for, as far as I recall this entire year. now there is one asterisk to that. In usually the first six hours of launching a new model run, something breaks because you set a config wrong, you made a small mistake, et cetera. So that's usually there's a little bit of intervention, but that's always within like call periods, right? Not on call. And I think that's starting to now compound. So the model we're releasing now, I love it. It's amazing, but we're already onto the next one. and I think that's the way it should be.Laguna, Five-Week Builds, and Zero On-Call EventsVibhu [00:22:50]: Hey, I also just wanna point out, so for context, this was like a month ago. we found it in the tech report, so we just came in with, “Okay, new model's dropped. Haven't heard about it.” We wereEiso Kant [00:23:02]: Yeah, we're very used to doing this every few months.Vibhu [00:23:03]: We're, we're very much like, “ okay, look, it's like, on par with Kimi, DeepSeek, whatnot, the small ones, Gemma level. Oh, it's a very cool paper on what goes into building.” And then we hit this page, right? Like literally page two of tech report is, “This process allowed us to build the small model from scratch to delivery within five weeks applying the lessons”. And then I'm like, oh, this paper is not about here's a tech report of benchmarks and here's how many tokens it was trained on. Like for people that wanna dive more from what we're not gonna discuss on the podcast, it's all laid out here, right? FromEiso Kant [00:23:38]: YeahVibhu [00:23:39]: Custom software that agents can use to interface with training code, training data.Eiso Kant [00:23:45]: Yeah. Well, link the paper correctly, so yeah.Vibhu [00:23:47]: Yeah. All that stuff. read the paper here, but,Technical Report Principles and Streaming Training DataEiso Kant [00:23:50]: But I would like to. I love principles, and I think that is a good starting off point for maybe telling some stories. Maybe we can go one by one past the principles. I'll just call out that Dagster just got bought by a Prefect.Vibhu [00:24:01]: Yeah.Eiso Kant [00:24:01]: Isn't it fun? But yes, I'm very familiar with Dagster. just anything where like they trigger some story.Vibhu [00:24:07]: So, well, I would say, well, experiments code's obvious, but I think one of my favorite things is, I don't know where it is in here, but early on, and I still think this is the case a lot of foundation model companies, people prepare their training data sets, they get packaged up, then they get copied over to a training cluster distributed across all of the nodes, and then training starts.Vibhu [00:24:30]: And we looked at this like three years ago and we were like That makes no senseEiso Kant [00:24:36]: You lose so much time because the moment you have to rematerialize the data set, you have to make a change, you have to fix something, et cetera, you've got all this time of like repackaging it, right? Toca- tokenizing it, repacking it, moving it over to a cluster, then distributing it across the nodes. The bigger your clusters are, you start using fancy like torrent-like algorithms to like distribute your data. So why aren't we streaming data into training? Right? Something that's very common and like just basicVibhu [00:25:00]: Like just in timeEiso Kant [00:25:01]: Just in time, like good computer science like principle. And that was one of the first things that I think unlocked - the model factory. Because the moment you start thinking about, well, a training job, it doesn't matter if it's a big hero run or a small like, post-training experiment, consumes a certain number of tokens per second, right? And it's not a lot, right? From a like a data, moving data perspective. So we said, well, we have our training cluster, and then we've got like our AWS kinda setup where we can build these amazing big data pipelines. We can set things up. We use Spark underneath the hood, like all these things.Vibhu [00:25:36]: But when you say AWS, it's not actual AWS, it's your internal AWS.Eiso Kant [00:25:39]: It's our internal-- No, it's our internal like just running like our infrastructureVibhu [00:25:42]: Site web servicesEiso Kant [00:25:43]: Exactly. Our stuff running on like an AWS account or on like any hardware, right?Vibhu [00:25:47]: Yeah.Eiso Kant [00:25:48]: And so once we made that shift into I can stream data into training, all of a sudden you realize a lot of things unlock. Because now you don't have to wait for the whole data set to materialize.Immutable Data, Experiments as Code, and Scientific RigorEiso Kant [00:26:00]: You now all of a sudden when you're running data experiments about mixing data, it's a config. Because you've got these data sources that are coming in, and you just - we have this service called Blender that's in the report, where we then say, “Okay, for this run, I want 20% of this source, 10% of this source. I want this much, so many epochs of repetition. I want this to be, shuffled in a certain way,” and your training job can start while the rest of the data is even still materializing. also what it does is because all of this underneath-- So for us, we treated the data layer underneath as like an immutable data layer, and that was really important. Like experiments as code, immutable data layer means that you can always go back and understand literally down to the single token at which cursor it went in on which version of the code.Vibhu [00:26:47]: Yeah.Eiso Kant [00:26:48]: And it took us a I have to admit, like the first year of Poolside, we understood that engineering had to get great, But we didn't understand yet, that this is ultimately in support of like a good rigorous scientific progress. We were quite a - We were a very small number of people, so a lot of it was YOLO ideas and YOLO runs.Vibhu [00:27:08]: Yeah.Eiso Kant [00:27:09]: And we built great infra for the YOLO runs. But once we realized that we treated data as immutable and code as always versioned, and you could always track and trace every experiment end to end perfectly, you could repeat everything perfectly, right? You have perfect reproducibility. I can still reproduce runs from two years ago if I wanted to, right? It enables the scientific progress, like the scientific process, and I think that took us probably about a year and a half into the company to figure out. We also had some great hires, like our head of applied research, Nikolai, who joined us from Yandex, who'd been working on language models since like the early 2020s, I think brought that into the company of like, “Hey, we wanna have even more rigor.” And then once we kinda had the combination of like increasingly more capable platform that allowed people to do more, but had this immutability, we were able to start “Okay, every experiment is truly an ablation. We truly need to understand it.” And I think we became much more scientifically rigorous in the last couple of years, and the infra underneath enabled it. and then there's just fun stuff like, andVibhu [00:28:16]: Yeah, a lot of it's fun, like even just the, one, you share all the ablations, two, picking the data sets, right? There's like a random small paragraph in here where it's just like, “Oh yeah, training data, we have some, we have an auto mixer.” it trains eight small models, scales them up, picks the training data set. We don't even need to look at it. I'm like, “Wow, a lot of engineering rigor there.” And there's just, there's just a lot in here.Publishing Research and Giving BackEiso Kant [00:28:40]: Yeah, and it'- and look, and we wanna put out more. Like we, We treat writing papers as something that we haven't earned the right for yet for a long time. So you earn the right to spend time, publishing research once you're at the frontier, because until then, you're catching up, and every minute and hour in this industry matters. Like I obsess over it, not just the wall clock time from idea to result, but just general like time every day that we, waste is one that doesn't allow us to catch up. But in this case, we said, “Okay, we're gonna give ourselves.” I think we gave the team like three or four days while still doing their work, like give everything in there. And to your point earlier, if your stuff, it's easy to like put it out. And so there's so many more things that we wanna talk about over time, and we will definitely start doing. And as we earn more of the right, but also now have like added to our mission that we want more foundation model companies to exist, you'll see us like be way more proactive, and just trying to keep dropping some of those like things that we've learned along the way that can help others like speed up.Vibhu [00:29:40]: Which is the other cool side of this, right? It's, it's not like, back to your point, it's not just here's the benchmarks of our training. If you want to replicate, here's experiments of optimizers, data sets, post-training. you lay out a lot of it here alongside here's your system for how to do it? So it's, it's really like promotingEiso Kant [00:29:59]: No, thank youVibhu [00:29:59]: Other people can do the same.Eiso Kant [00:30:00]: And by the way, I also wanna make clear, right, we have been incredible-- Like we've taken a lot of advantage of the fact of all the open research that others have published, Right? And you mentioned, the Chinese labs, and we I think it's important that there's, from every country and every culture and background, including like Western companies like us, there's different models that come out that people can choose to trust. But I think we do have to give credit where credit's due, right? The incredible Chinese lab have done an amazing job at sharing their research, and we have definitely like been on the receiving end of taking advantage of that. So when you're on the receiving end of something coming to you, I think it's, you also have an obligation to give back.Swyx [00:30:39]: Do you have a favorite or underrated Chinese lab that you wanna shout out? Everyone shout outs DeepSeek.Chinese Labs, Zhipu, and PersistenceEiso Kant [00:30:44]: That's a good question.Swyx [00:30:45]: Moaan obviously for Therapsi. Yeah.Eiso Kant [00:30:48]: Yeah, look, I think, I think obviously everyone's been talking about Zhipu lately, with 5.2. I think what most people don't realize is when they started.Swyx [00:30:59]: Yeah.Eiso Kant [00:30:59]: Right? They started years before ChatGPT.Swyx [00:31:02]: They just rebranded. YeahEiso Kant [00:31:03]: And so, I've like, I remember how hard it was to work on these things Before the rest of the world got excited about it. And so I have an immense amount of respect for people, who were working on improving models when it wasn't the sexy thing to do, when believing in LLMs, was gonna get you ridiculed. I remember like back in 2016 when we were doing what we'd call, machine learning on code with some of these models. we would-- people would just laugh at us, like they'd be like, “This makes no sense. Like why are you wasting all these, like, millions of dollars on trying to figure this out?” And so I would say they're probably the one that, I think deserves a shout-out, not just because their latest model is very good, but because they fought to get here. And I think, I think every foundation model company it takes time to get here, right? It took us three years to get to the model that we're, that we're now gonna be releasing. and now the time in between the models is coming, is counted in weeks. It's no longer counted in months or years. But this stuff's hard. and if we can make it a little bit easier for the next person, like we should all do so. Because if we don't do so, we're, we've got a small window before models are really impacting recursive self-improvement to a level where catching up otherwise might become unfeasible. And we should try to, in that window, encourage as many labs or however we wanna call them, like to start. And so one of my currentEiso Kant [00:32:36]: Mission, but qualm is like I wanna encourage whoever is a researcher right now who thinks they can tackle this to go and leave and become my competitor.Eiso Kant [00:32:45]: Like start another foundation model company because I think we need it. I think otherwise we're not gonna be in the world where, I don't want to just be the fifth or the sixth company that wins. I wanna look at a world where there's lots of choice.Starting a Foundation Model CompanyVibhu [00:32:57]: What else do people not see in starting a foundation model? it's, there's a lot of compute, there's a lot of capital required, a lot of compute. You lay out model factory and how to do the training, but there's a lot there, right? That's,Eiso Kant [00:33:10]: Well, look, it's, I in turn-- this is an oversimplification, and I always asterisk it with that because it can land a little bit the wrong way in people's minds. But I think you can sum down, And I saw it, 95% of model building to just doing, you're just doing two things. You're improving data or you're improving compute efficiency. And I know that feels like an oversimplification for the incredible, like, Gifted and skilled work people do. But if you really look at it, like what are we doing? We are looking at data, we're generating new data, we're improving data. and the only way to do that is to look at the data, right? That's a big part of foundation model building. And on the other hand, we come up with these incredible breakthroughs in inference, in architecture, and new attention mechanisms. But what are they really doing? They're bringing compute efficiency. Now, we have definitely had some breakthroughs over the years that allow for more model capabilities. But at the limit, if you could train a large enough model, right, like, and you had infinite compute, we probably-- if you had infinite compute, you'd be at AGI probably already tomorrow.Eiso Kant [00:34:12]: Right? Like it's not. And so, and let me say that infinite compute with infinite ability of much faster networking because networking ends up being more of the bottleneck than compute. But, so I do think that's, those are the main things. And to just realize that this is engineering. I think it's become more obvious, but I think for quite a few years, people have held foundation model companies and researchers and others on this pedestal of like you're doing incredible magic or rocket science, or only like, Nobel laureate physicists can do this. And don't get me wrong, there are some really hard problems that need to be solved, but a lot of the work that all of us are doing on a day Is not sitting down trying to solve a math theorem. A lot of the work that we're doing is just really doing the basics right, writing good code, looking at data, improving it, running experiments, looking at plots, trying to see like, hey, trying to shape our intuitions. And a lot more people could be highly capable researchers. and I think that's, it feels far for people to do so. But I've seen in our own company, we've seen engineers become researchers because the model factory allowed them to be, have a much lower hurdle of running experiments and trying things. And one of the guys on our team who started as an engineer building our agents is a legit reinforcement learning researcher now, making real progress. and that happened in the span of like six months. that would've not been what I think most people assumed was possible, a couple of years ago.Swyx [00:35:46]: Yeah. I think one of the interesting moments is when you can self-host, like, if in a programming language, like if you can compile the language in the language, the equivalent is can you use your own tools, right? You have the pool CLI, you have your own models. presumably you're not only using your own models. There's no way. But like, what's that percentage over time?Laguna S, Persistence, and Behavioral GainsEiso Kant [00:36:10]: This is the first model that we're releasing that is starting to meaningfully contribute to our own work. It's not a it's not state-art model yet. Fable and other, they're, they're very capable models, but Laguna S Is really interesting. I'm gonna pull up the quote. Peng Ming, one of our heads of applied research, said something, last week as the model came out about 10 days ago, much better than we had hoped for or expected. And he said, I have the feeling that a lot of the gains in Laguna S come not from more intelligence, but more from different behavior, more verification, less taking things for granted, not declaring victory early, and being way more persistent. And to be honest, those are more predictive than raw intelligence for success in human also to some degree. And this was, he wrote me this on 5th of July on a Sunday, and it's been burned in my brain ever since because the Laguna S model, as you'll see it and why it does so well on benchmarks and why it does so well in using it on a day basis, is that it's just incredibly persistent. It reasons a lot. I do call that out. We have work to do on making it more efficient. We have to work to do on offering different reasoning modes. But this is the model that has been able to do things that I never thought it could do. A hundred eighteen billion 8B active model, which is not that large. It fits on a DGX Spark and still runs at, thirty, forty tokens a second on a Spark, is able to solve Erdős 397 independently. It's able to do complex programming tasks. It's able to. I asked it this morning to make me a Fi scanner without using any external libraries on my Mac, and it's, like, figuring out, like, the core WLAN API by really persistently trying to understand it without access to the internet. And more, I love vibe checking. I've probably spent eight to ten hours a day with this model for the last ten days.Eiso Kant [00:38:05]: I'm not exaggerating. I was on my eleven-hour flight yesterday. I spent ten hours reading trajectories and traces and, like, of the model.Eiso Kant [00:38:12]: And what I take away from it is exactly what Peng Ming said. We are gonna be able to squeeze so much more out of smaller models than I think we had imagined in the industry because, yes, there's intelligence and larger models are more intelligent. Like, no doubt about it. We should continue to scale up. but the behaviors of being really persistent, of being able to backtrack when you're wrong, of, like, understanding how to interact with your environment show us that we can get a lot more out of it. And this, for me, has created a bit of a Question in my mind the last couple of days. If you think about where we're using models today, right? We are using models, say, for knowledge work. Represents twenty-five percent of the global economy, twenty-five trillion dollars of work.Eiso Kant [00:39:00]: As we scale up models and they become more intelligent, we are excited about using them more and more for pushing the frontier of science.Small Models, Knowledge Work, and CommoditizationEiso Kant [00:39:08]: And if you look at the frontier of science, like true breakthroughs in science, they have been linked, they are linked to more intelligence in many places. Einstein figuring out general relativity is able to bring ideas together that other people would have not brought together. And I think one of the many dimensions of intelligence is the ability to do that, and it's something we clearly see that as models get larger and more capable, they're able to pull more ideas and threads together that a smaller model wouldn't be able to.Eiso Kant [00:39:36]: And we're starting to see examples of that in medicine and, like, in bio and other things. But if you think about the majority of knowledge work that we do, and it includes building software. I'm a software developer at heart first and foremost probably, although I probably can't say it that much anymore as I don't write production code in years, is that what makes us good is our persistence. It's our ability to encounter a problem and backtrack and say, “I need to go figure out this bug. I need to go research this. I need to go look at the documentation. I need to, like, try different, five different ways to see, like, if I can solve it.” But it is not necessarily bringing three ideas together from radically different fields. And so if we are now seeing, and I think Laguna S is an example, that we are able to make a relatively small model much more capable than I had definitely predicted or any previous, like, benchmarks had shown for any model remotely this size or even larger, At least on coding tasks, that it's because of the behaviors. And so now the question I have, and I don't have an answer, it is I know at the limit, so infinite model size, right, extremely large model, and the cost of that model is gonna be very expensive to run. We know this, right? So larger model ROI.Eiso Kant [00:40:52]: So I know that at the very limit, I'm not gonna use the world's largest model one day, quadrillion parameter, whatever crazy, like, scale we scale up, to do a basic coding task. Already today, I'm starting to size down for certain tasks.Eiso Kant [00:41:07]: So it means that there is an optimal. It means there's some curve that goes as we go up to model size for knowledge work, at some point we're at the peak, and after that, the return on investment of using a bigger model, just doesn't make sense.Eiso Kant [00:41:22]: Now, I think the question is, before I would have thought that peak was extremely very far away.Eiso Kant [00:41:30]: This model for me is the first sign that Maybe that peak is At a trillion, five trillion, ten trillion. Maybe we can just squeeze way more out of these models. I'm no longer thinking that we need two or three orders of magnitude on the largest models to be able to, solve knowledge work, the accounting, the legal, the code that we write. And so if that holds true, It is an argument for the commoditization of models. It's an argument that open source can win and, like, succeed in this world. And now it's of course a self-serving argument and it's a hopeful argument, but theoretically at the limit it works. We just have to go discover in the next couple of years of how much more we can squeeze out. Now, I do want to put a big asterisk. This does not mean I'm against scaling models. I think we ultimately only succeed if we scale our models as large as our competition. I do not like. I think we should not put our head in the sand and say we're gonna be king of open source small models. I think that's, It's a out. It's trying to be king of your own kingdom, but not realizing what the rest of the world's doing. All of us rather use a smarter, faster, more model. It's a sign of hope. And so I don't wanna overly state this is a good model. We have a long way to go to get to the state-art. But what hopefully people take away when they use this model is that the behaviors inside of it are what push it to be far more capable, less than necessarily the number of parameters.Pre-Training, Mid-Training, and RL Moving EarlierVibhu [00:43:03]: Is that mostly post-training? LikeEiso Kant [00:43:05]: YesVibhu [00:43:05]: Right.Eiso Kant [00:43:06]: It's entirely post-training.Vibhu [00:43:08]: Are we done improving anything on training? Is, like, training done?Eiso Kant [00:43:12]: No.Vibhu [00:43:12]: Okay.Eiso Kant [00:43:13]: SoVibhu [00:43:13]: I just wanted to cover training, and then we go post-trainingEiso Kant [00:43:15]: Training is not done. I mean, look, there's a part of training of just dealing with skill, right? Every new order of magnitude of model skill, you are going to get new things you gotta solve for. That'- but those are ultimately, engineering challenges.Eiso Kant [00:43:31]: I have a, I would say, a not commonly held opinion that reinforcement learning Will move earlier and earlier into training.Vibhu [00:43:42]: Yeah, training.Eiso Kant [00:43:44]: Not even training. Like training today, right, is, like if you look at - So we've been working on this for years already. and I think the best-- I think the first time we saw it out in public was the DeepSeek Zero paper. this is a year and a half ago, I think, if I recall correctly. where, you can Very early on in a model as it starts capable of being able to use language, et cetera, induce reasoning. and so the question that I have is like, we have this- we have the dataset that's the web. and the web, I think we could arguably say probably has The totality of humanity's knowledge somewhere encoded in different places. It's a huge variance degree of quality, from garbage data, and like once you look at training data, you really get humbled of like what the web is, to like, the most greatest scientific papers and best blog posts and like, best transcripts and whatnot.Eiso Kant [00:44:39]: And so now What we are trying to figure out, and have been doing a lot of work on, and it's a place where maybe not as open as we're on other things, but we will become more over time. we've been spending a couple of years really doing research on how can we turn the web into not just next token prediction, but into a way to teach the model to think earlier in its training. and I think there's a huge amount of gold to be found there. I think we are right now in, we've got some drugs in the industry. One of the drugs is distillation. Another drug is, more environments. Like, and they're great, and they make us feel good, and they make the models better, and like we're all addicted to them, and we'll use them, right? in various different ways. and but ultimately, I think we are still barely squeezing out of the web what we should be getting out of the web.Eiso Kant [00:45:33]: I think just next token prediction during training is not enough.Eiso Kant [00:45:36]: AndVibhu [00:45:38]: YeahEiso Kant [00:45:38]: I think we'll see some very interesting things still happen. and that RL in post-training to induce behaviors, to improve things, like I think - the whole world knows how to do this now. I think we're, we're scaling it up. Everyone is. But I wonder if we need to go as far as we're going today with environments. I'm not sure yetVibhu [00:46:01]: You mean we're going too far?Eiso Kant [00:46:02]: I'm, I'm not sure if the path to AGI is justVibhu [00:46:06]: Is more environmentEiso Kant [00:46:07]: More environments.Vibhu [00:46:08]: It seems like a never-ending, “Okay, I want instruction manual for this table, right? Am I gonna environment out building furniture? Or are we just gonna tail end like we need some general solution?”Eiso Kant [00:46:19]: I think there is, I think there's an ability to generalize more from the web. but I also am very encouraged, like when I look at Laguna S and, which is post-training is, well, is the big impact there. and I see like, oh, wait a second, just by making some of these behaviors much better, we're able to get so much more out of it. It just changes a little bit the way you think about intelligence.Vibhu [00:46:40]: Yeah. The analogy people draw often is the RL phase is where you don't learn as much new knowledge. You shiftEiso Kant [00:46:46]: Yeah.Vibhu [00:46:46]: Yeah. So, you shift distribution, and you can have it reason towards what you want. on your point about training, a lot of training is still just continue training in a domain, say medicine, then you do RL. So still justEiso Kant [00:47:00]: It's just better data, right? Like, I mean, training, ooh, I like how we invented this word. Like it's effectively just like,Vibhu [00:47:06]: Second phaseEiso Kant [00:47:07]: It's the second phase of training With like a really dumb way to do a curriculum. But like ultimately, what you'd want is a curriculum from token zero to token 30 whatever or 40 trillion tokens that really truly is the optimal curriculum for the model to learn. But training is essentially a stage curriculum on the web because we do not have to compute, And, effectively to try to ablate the perfect curriculum, right? And so I'm pretty sure that you'll start to see people talking soon about some other term, and there's two or - ‘cause now we do this, right? We talk stage two and stage three and stage four training and like. But ultimately, all we're doing is we're trying to assign a curriculum to the web data that we have to allow the model to learn better. I think at some point, as things get compute, as models get cheaper to run, as the next generations of compute, this will become more of a continuous spectrum. I also think the reason, by the way, you have training and like stage two and stage three is organizational, Right? It'- this is, I think, a thing where-- that we really try to avoid with the model factory is like Training exists because there's a training team now, right? There's people, or like people in training decide to focus on like a training effort. but what you really want is engineering and scale of experiments that allows for a much more continuous spectrum that you don't, you have infinite stages. Now, we're not there. Compute's not there. Organization design is not there for it yet. but I think we'll get there. we'll look back on a couple of years and be like, “Oh my God, it was so cute that we did our training data like this in such a like naïve way. Like we barely ordered it. We didn't really do a good job at likeCurriculum, Auto Research, and New ObjectivesVibhu [00:48:48]: The building that curriculum will get you that in the industry.Eiso Kant [00:48:51]: And I'll confirm that, when I talk to some researchers that this is a lot of the focus now is like how does training change and what is the next objective other than, next token prediction. I assume you don't have the answers, but you have some ideas.Vibhu [00:49:02]: We have some ideas. We're not ready to talk about it yet.Eiso Kant [00:49:05]: Yeah.Vibhu [00:49:05]: We've been working on them for years, and I think that's the one thing that's also like you asked earlier about, like what's not obvious about building a foundation model company is that you are constantly balancing the table stakes work, the recipe worksEiso Kant [00:49:19]: Yeah.Vibhu [00:49:19]: Versus like your, my crazyEiso Kant [00:49:22]: Pure researchVibhu [00:49:22]: Breakthrough.Eiso Kant [00:49:22]: Yeah.Vibhu [00:49:22]: Pure research and finding that balance and adjusting the percentage to it based on where you are in the race is really important.Eiso Kant [00:49:31]: I mean, so like, this is a nice way. I was gonna bring up auto research at some pointVibhu [00:49:35]: YesEiso Kant [00:49:35]: As another Andrej invention, or coinage, which is like, I honestly, like how many objective functions can there be, right? Like just try 1,000 of them, set it running, whatever.Vibhu [00:49:47]: Man, it's alsoEiso Kant [00:49:48]: Like what you're looking for. You're looking for loss curves like that, likeVibhu [00:49:51]: It's also a thing people take bets on, right? When you say more Neo labs, you're doing a version of we'll do foundation models, scale them up, next token predictors. A lot of other Neo labs that we see want to take a completely different approach, right? At some level, you're right. It's all, compute efficiency, and that's the net objective. But some are okay, different architecture, like vastly different amounts of compute spend. So some are different. They're not justEiso Kant [00:50:19]: YeahVibhu [00:50:19]: They're like, 99% not balancing, here's the vanilla and scale up. They're 99% on, here's novel research that'll change everything.Eiso Kant [00:50:27]: And I think, Luke, I think you. It depends when you started as well, right?Pure Research vs. Table StakesVibhu [00:50:30]: Yeah.Eiso Kant [00:50:30]: When we started, like the novel thing we did was reinforcement learning on code. No long- that's no longer novel by far, but we were like, - that's where we obsessed over when no one believed in RL. So you have to when you start the company, you have to have your own idea. You have to have something that's different that allows you to speed up, right? For us, it was RL to LLMs that later became common, like, Knowledge. But in the beginning, it wasn'tVibhu [00:50:53]: It's cool. this was like your original 2023 blogEiso Kant [00:50:57]: YeahVibhu [00:50:57]: Of purpose.Eiso Kant [00:50:58]: Yeah.Vibhu [00:50:59]: And like you do lay it all out here.Eiso Kant [00:51:01]: We laidVibhu [00:51:01]: The blog is pretty underrated, right? The whole RL on code was very early on.Eiso Kant [00:51:06]: Very early. And even we had to argue with people, like we say here things like to push beyond current capability, to train your own foundation model. We had to argue with people that it mattered that you had your own like, base model. you can fine-tune your way to success, right? major capabilities emerge from training a base model made accurate and useful during fine-tuning.Vibhu [00:51:23]: Which like, for perspective at the time, we knew closed models, OpenAI, Anthropic were huge. The open models we had were like Mistral 7B, a 30B, a 70B.Eiso Kant [00:51:35]: When weVibhu [00:51:35]: YeahEiso Kant [00:51:36]: The date on this thing is wrong. When we published this, it was April 2023. I think this was justVibhu [00:51:42]: YeahEiso Kant [00:51:42]: Happened on a migration, probably found it on archive.org.Vibhu [00:51:45]: Mistral.Eiso Kant [00:51:46]: Mistral had started, we started on the same month, right?Vibhu [00:51:49]: Yeah.Eiso Kant [00:51:49]: So this wasn't even, there was only, I think, Llama out at the timeVibhu [00:51:52]: SnellEiso Kant [00:51:52]: And that's it, right? And so, but I agree. I think we wan

This Day in Legal History: The Senate Rejects Court-PackingOn July 22, 1937, the United States Senate rejected President Franklin D. Roosevelt's plan to reorganize—critics said “pack”—the Supreme Court, voting 70 to 20 to send the bill to a quiet death. It was a stinging defeat for a president at the height of his popularity, delivered by his own party, and it settled a constitutional question that still shapes how we think about the independence of the judiciary.The background was a collision between the New Deal and the Court. Through the mid-1930s, a conservative majority on the Supreme Court had struck down key pieces of Roosevelt's economic program as unconstitutional. Frustrated after his landslide 1936 reelection, FDR proposed legislation that would have let him appoint a new justice for every sitting justice over the age of seventy—which, not coincidentally, would have allowed him to add up to six new justices and swamp the opposition. He framed it as a matter of efficiency and helping overworked elderly judges, but nobody was fooled; it was a naked attempt to change the Court's decisions by changing its membership.The plan backfired, and the reasons are the lesson. Even senators who supported the New Deal recoiled at the precedent—if this president could enlarge the Court to get the rulings he wanted, so could the next one, and the Court's independence would become a fiction. Meanwhile, the Court itself defused the crisis: in the spring of 1937, Justice Owen Roberts began voting to uphold New Deal legislation, the famous “switch in time that saved nine,” which took some of the urgency out of FDR's demand. The significance of July 22, 1937 is that it established a durable, if unwritten, constitutional norm—that the size of the Supreme Court is essentially off-limits as a tool for a president to overpower rulings he dislikes. The number nine isn't in the Constitution, but the bipartisan rebuke of court-packing helped make it feel almost as if it were.An analysis of the closely watched lawsuit by Meta employees over AI-driven layoffs highlights a hard truth: even when workers suspect an algorithm decided their fate, proving it is enormously difficult. To recap, 26 current and former Meta employees sued, alleging the company's internal AI tools flagged them for termination because they have disabilities or took protected medical, parental, or family leave. Their theory is mechanically specific: because tools like the “Metamate” system scored employees partly on data such as keystroke activity, workers who were lawfully out on leave generated fewer data points and were disproportionately ranked as low-value. Meta cut roughly 8,000 people—about ten percent of its workforce—and says humans, not machines, made the decisions. Here's why these cases are so hard to win. Anti-discrimination law generally requires the worker to show the employer's decision was tainted by a protected characteristic, but the employee usually has almost no visibility into how the AI actually worked—the models, the training data, and the weighting are the company's closely held secrets. On top of that, many employees have signed arbitration agreements, funneling their claims out of open court and into a private process that's harder to see into and to appeal. The significance is that this appears to be the first case of its kind against a major U.S. company, and it exposes a growing gap: as employers hand more consequential decisions to opaque algorithms, the legal tools workers have to challenge those decisions—built for an era of human managers—may not be up to the job of proving what the machine did.Analysis: Meta employees' lawsuit shows that if AI fires you, proving it is the hard part | ReutersA split panel of the D.C. Circuit has struck down a long-standing National Labor Relations Board doctrine that protected unions after a business changes hands, ruling that it conflicts with federal labor law. The doctrine at issue is the “successor bar,” and it works like this: when a company is acquired and a new employer takes over, that employer generally cannot challenge or withdraw recognition from the existing union for a reasonable period—about six months—giving the union and workers a window of stability to bargain with their new boss. The court held that this Board-created rule isn't consistent with the National Labor Relations Act. What makes this ruling bigger than one labor doctrine is the tool the court used to get there. The decision applies the Supreme Court's 2024 Loper Bright ruling, which overturned the decades-old Chevron doctrine and ended the requirement that courts defer to a federal agency's reasonable interpretation of an ambiguous statute. Without that deference, the D.C. Circuit felt free to substitute its own reading of the labor law for the NLRB's. This is exactly the dynamic I wrote about in my Bloomberg column last week in the tax context—the death of Chevron doesn't erase statutory ambiguity, it just moves the power to resolve it from agencies to courts. The significance is that we're now watching that shift play out across the administrative state: settled agency doctrines, some decades old, are suddenly vulnerable to being reinterpreted by judges, and here the immediate losers are unions and the workers who counted on a bargaining foothold after a merger.US court says longstanding NLRB rule on post-merger union bargaining is invalid | ReutersA federal judge has temporarily blocked the administration from stripping work authorization from tens of thousands of asylum seekers and immigrants with Temporary Protected Status. U.S. District Judge Nathaniel Gorton in Boston sided with a coalition of immigrant-rights groups and labor unions, halting U.S. Citizenship and Immigration Services from moving ahead with a set of policies while he weighs a longer-term pause; he said he'll rule by August 5. Here's the stakes and the legal frame. A work permit—formally, an employment authorization document—is what lets many immigrants lawfully hold a job while their asylum case or protected status is pending. Yanking it doesn't just threaten deportation down the line; it immediately jeopardizes people's livelihoods and their employers' workforces. The contested policies were designed to implement immigration restrictions Congress enacted last year as part of the administration's signature tax-and-spending law, the One Big Beautiful Bill Act. The plaintiffs argue USCIS is implementing those provisions in ways that exceed what the law allows and skip required procedures. A temporary block like this one preserves the status quo—keeping people employed—while the court decides whether the government followed the rules. The significance connects to a theme we keep returning to: courts serving as a check on how fast and how far the executive can move in reshaping immigration, insisting that even policies rooted in a real act of Congress still have to be implemented lawfully and with proper process.US judge blocks Trump administration from stripping immigrants of work permits | ReutersAnd finally, in my column for Bloomberg Tax this week, I dig into a self-inflicted mess in California: lawmakers scrambling to rework a business tax-credit cap that they apparently didn't realize would kneecap Hollywood film studios. My core argument is that California is directionally right to resist subsidy bidding wars, but wrong to rewrite the economics of credits it has already issued after companies have started relying on them.Here's what happened. Since 2024, California has capped the total tax reduction a business can take from all its credits at $5 million a year. That cap was set to expire after 2026—right as productions were going to start claiming credits under a newly expanded film incentive the state had just touted as a centerpiece of keeping film jobs in California. Instead, a bill called SB 122 extended the $5 million limit through 2029 and then converts it to the greater of $5 million or 70% of taxes owed. The part that really gets me is the admission underneath it: lawmakers passed a $351.7 billion budget without apparently understanding how this cap would interact with the film credit they'd just enlarged. As one assemblymember candidly put it, “I'm not sure who knew what about what.” It looks like the cap was really aimed at large research-and-development credit stockpiles, and film credits just got caught in the crossfire.My argument is that the distinction between prospective and retroactive matters enormously here. It's one thing for California to decide, going forward, that future subsidies will be smaller or conditioned—that's legitimate fiscal discipline, and I don't think Hollywood should get to dictate tax policy just by threatening to decamp to Georgia. But it's another thing entirely to change the timing and practical value of credits after studios have already committed workers, facilities, and financing in reliance on the old rules. When a state does that, it makes itself a less credible counterparty, and it quietly reduces the value of every future incentive it offers, because businesses will start discounting California's promises for legislative risk. So my prescription is targeted: protect the film credits already awarded under the prior rules, keep a real limit on the big accumulated R&D credits that were the actual target, and replace the blunt across-the-board cap with rules tailored to how these very different credits actually work. California doesn't have to choose between fiscal discipline and keeping its word—its tax policy can be skeptical, but its promises should still mean something.California's Business Tax Credit Cap Needs More Targeted Changes | Bloomberg Tax This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.minimumcomp.com/subscribe

Thirtieth birthdays are among the most reflective celebrations that we have. Born out of one of the largest corporate mergers in history, Novartis was founded in 1996 and celebrated its 30th anniversary this year. As such, the Swiss pharma giant is reflecting on its corporate branding down to the color palette as it eyes a future where its R&D operations lead to new drugs on the market targeting a host of disease states and medical conditions. In order to articulate its mission and underscore how it has evolved from a marketing perspective, Novartis is taking a look back before it sets its eyes on the future. This week's episode features reporter Bella Czajkowski in conversation with Novartis' chief corporate affairs officer Michelle Weese. The pair discuss the history of Novartis, how it is recognizing its 30 years in business and what we can expect from the drugmaker as it relates to marketing in the months and years ahead. And for our Trends segment, executive editor Jack O'Brien is joined by editor-at-large Steve Madden and pharma editor Lecia Bushak to talk through the FDA's proposed de facto pharma ad ban now that the dust has settled — at least a little bit.  Check us out at: mmm-online.com Follow us: YouTube: @MMM-onlineTikTok: @MMMnewsInstagram: @MMMnewsonlineTwitter/X: @MMMnewsLinkedIn: MM+M To read more of the most timely, balanced and original reporting in medical marketing, subscribe here.Music: “Deep Reflection” by DP and Triple Scoop Music. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

How do you build something nobody asked for—and turn it into an overnight obsession? In this episode of the Innovation Storytellers Show, I speak with Ross Richardson, Chief Design Officer at SharkNinja, about the thinking, testing, storytelling, and consumer obsession behind some of the company's most successful products. Ross leads a global team spanning mechanical and product engineering, industrial design, product innovation, and R&D. But his path into product design began much earlier, from taking things apart and borrowing his father's tools as a child to studying product design engineering and building a career around understanding how people interact with physical products. Our conversation reveals how SharkNinja approaches innovation by starting with consumer behavior rather than technology for technology's sake. Ross explains how the company identifies what he describes as "desirability gaps," finding experiences people already want but cannot easily recreate at home. From frozen drinks and ice cream to coffee and personal care, the opportunity often comes from making something previously difficult, intimidating, or inaccessible feel simple and repeatable. We also discuss the story behind products such as BlendBoss and how observing changing consumer behavior can lead to unexpected opportunities within established categories. Rather than assuming innovation always requires inventing an entirely new market, SharkNinja watches how people live, what they carry, what they share on social media, and how cultural trends are changing their expectations. At the center of this process is a deceptively simple challenge: "prove it." Ross explains why new ideas must withstand technical testing, commercial scrutiny, consumer validation, and internal skepticism before moving forward. He also takes me inside SharkNinja's show-and-tell culture, weekly product demonstrations, in-home user trials, observational research, physical prototyping, and its "red team" approach to bringing fresh perspectives into the development process. In an era where AI can help almost anyone create a polished presentation or convincing visualization, Ross argues that evidence matters more than ever. A compelling story may earn attention, but data, testing, consumer insight, and a working proof of concept are what help turn skepticism into stakeholder buy-in. We also discuss why teams need room to dream before inviting skeptics into the process, how collective enthusiasm can help engineers solve seemingly impossible problems, and why admitting that a favored idea is wrong can sometimes lead to a far better product. As AI changes how we research markets, generate concepts, and communicate innovation, my conversation with Ross is a timely reminder that understanding human behavior remains at the heart of successful product development. The tools available to innovators may be changing rapidly, but our ability to listen to consumers, identify genuine needs, test assumptions, communicate breakthrough ideas, and prove that an idea deserves to exist still determines whether innovation connects with the people it was designed to serve. How can we combine faster, technology-enabled innovation with the deep consumer understanding needed to create products people genuinely want? Listen to the full conversation and share your thoughts.  

What if the fastest way to grow your business isn't more clients, but fewer? Today, Damien Schreurs founder of EasyTECH, shares how he built his side hustle: testing ideas for free, negotiating time away from a full-time R&D engineering job, and running on minimal sleep just to make it all fit. When he finally went all-in, he was confident he had the right plan. He didn't. Damien came within one late invoice of bankruptcy, and it took a friend from his networking group pointing him toward the right training and mentorship before he found the number that actually mattered: revenue per customer. That single shift led him to cut his client list from 45 down to 9, multiplying his revenue per customer by nine times in the process. In this episode, you'll hear: The real math behind "fewer clients, more revenue," and why total client count can be the wrong thing to chase How Damien nearly lost everything before he found the metric that turned his business around Why he kept running a newsletter for years out of pure habit, long after it stopped working for him The one regret he still carries about staying quiet about his business, and what he'd do differently How he's using AI today as a co-CEO, CMO, and CFO for EasyTECH Do you like what you're hearing? Consider giving it a caffeinated thumbs up. We'd really appreciate it! Need a little (and sometimes big) push to start and stay focused to grow your side hustle? Dive into my online Masterclass: How To Turn Your Thoughts Into Wanted Things. For the full show notes head on over to the home of Side Hustle Hero. https://www.sidehustlehero.com/199 Connect with Damien: LinkedIn Macpreneur Connect with Joan: Instagram Facebook About Joan Be on the show! Tell us about your side hustle success story!    

Unlock the future of plant-based proteins with PURIS VP of R&D, Kushal Shandak. In this episode, discover how a 40-year journey from soy to peas is reshaping the alternative protein industry—and learn the secret to overcoming technical challenges like solubility, allergen safety, and acidic stability. Shandak shares how Purist's innovative, vertically integrated model is making pea protein more accessible, sustainable, and tastier than ever, paving the way for the next wave of consumer-loved plant-based foods.

Wissal Ben Moussa a cofondé Sand to Green en 2022, une start-up qui s'appuie sur des technologies de pointe et s'inspire de la nature pour proposer aux paysans des solutions plus durables et mieux adaptées au changement climatique. Des pratiques qui permettent de régénérer les sols quand l'agriculture conventionnelle les appauvrit.  De notre envoyé spécial de retour de Guelmim, Une ferme à la végétation luxuriante, c'est l'exploitation familiale de Wissal Ben Moussa à Guelmim, dans le sud du Maroc, à la lisière du désert. « Je suis de formation ingénieure en agroalimentaire. Je bossais dans une multinationale, raconte-t-elle, j'étais responsable R&D, on formulait des cubes de bouillon et on devait utiliser des colorants, etc., que tout le monde jugeait assez toxiques. » Son attirance pour le vivant et la recherche d'un travail qui a du sens vont la conduire à changer de voie. « Donc, grosse prise de conscience écologique personnelle et d'une manière tout à fait automatique, je me suis dit que si j'avais un impact à faire quelque part, c'était vraiment au tout début de la filière, sur la partie agro, précise Wissal Ben Moussa, sauf que j'avais zéro connaissance en agro. » À lire aussiFemmes de la terre: Agathe Vanié, une combattante de l'agriculture biologique en Côte d'Ivoire [2/5] « On a vraiment démarré du terrain » Elle commence par un petit potager de 200 mètres carrés et découvre la permaculture et l'agroécologie. Sa ferme est un véritable jardin d'Éden. Pourtant, Wissal Ben Moussa est partie d'une terre aride, inhospitalière en apparence. Elle est parvenue à y semer de nouveau la vie. C'est ici qu'elle mène des expérimentations pour la start-up Sand to Green. « L'agriculture de régénération est la solution à la crise que l'agriculture est en train de vivre aujourd'hui face au changement climatique. C'est de là que l'idée de créer Sand to Green est née, explique l'ingénieure. L'idée de créer une expertise, une solution qui soit ancrée dans la terre. On n'a pas démarré dans des laboratoires, on a vraiment démarré du terrain. » Toutes les données du sol et des cultures sont collectées pour alimenter une plateforme qui permettra une standardisation des bonnes pratiques. Les agriculteurs du monde entier pourront ensuite les appliquer. « Il y a un mot en anglais que j'adore pour décrire un fermier, ils vont dire "land steward", du mot "stewardship", précise-t-elle. En tant qu'agriculteur, tu es juste là pour prendre soin de ce terrain-là. Tu fais partie de tout cet écosystème. » Aujourd'hui, la start-up Sand to Green accompagne des milliers d'agriculteurs à travers tout le continent, en Côte d'Ivoire, en Tanzanie ou encore en RDC.  À lire aussiFemmes de la terre: à Soweto, une ferme s'inspire de Nelson Mandela [1/5]

This week's video transcript summary is here. You can click on any bulleted section to see the actual transcript. Thanks to Granola for its software.EditorialIntelligence: Who Owns it?This week the word “AI” feels too small.AI is a technology. Intelligence is its product. And if intelligence is the product, the question is no longer just: Which model is best? Who has the cheapest tokens? Who owns the weights? Who controls the data center? Those are important questions, but they are lower in the stack.The bigger question is simpler and more political:Who owns intelligence?That sounds abstract until you make it concrete. Intelligence is becoming something companies can capture, package, serve, meter, route, improve, and sell.It can write code, answer questions, design molecules, automate offices, run agents, draft legal work, advise scientists, serve consumers, and reshape workflows. It is not merely software. It is a general-purpose capability. And all humans could benefit from more of it.General-purpose capabilities have a habit of becoming public questions. But the default answer, that public good is best delivered by government, is the wrong answer in this context.The Product Is IntelligenceWe should stop talking about AI as a feature and start talking about intelligence as the universal thing that is delivered as an input to the world.Water is an input. Electricity is an input. Literacy is an input. Connectivity is an input. Once a society depends on them, access stops being optional. Nobody needs government to build every well, power plant, school, or network. But everybody understands that a civilization cannot be organized around less than universal and reliable access to foundational inputs.Intelligence is reaching that level of importance now that we all know it is real.Government should not own it, operate it, or develop it. Quite the opposite. Companies are the right actors to build fast, compete hard, improve models, serve customers, and discover the real use cases. Self-interest is a useful framing here. Markets are good at finding demand, reducing costs, and turning invention into services people actually use.Companies are the right operators, developers, and owners. But that does not settle the real question of who owns the benefits. That is an economic question.If intelligence becomes metered infrastructure, what happens to the value it creates?The Ownership StackThis week's articles keep circling the same issue from different directions but in the nature of ‘circling' never quite nail it.Jamin Ball's “Own Your Weights” starts with the enterprise version of the question. Owning a model file is not enough. The durable asset is the loop: the data flywheel, the evaluations, the reinforcement system, the workflow learning, and the operating context that lets capability compound.Benedict Evans' “Ways to Think About Token Pricing” adds the market layer. Tokens may become essential, abundant, and cheap, like mobile data. But being essential does not guarantee that the token layer captures the value. The money may move up the stack to whoever owns the workflow, the customer, the distribution, or the application.Alex Karp's fight with the labs, reported in “Alex Karp Is Saying What Every Angry CEO Is Thinking About AI”, is the same argument in sharper enterprise language. Companies are afraid that model providers will not just sell intelligence, but learn from customer workflows and then move into the markets where those workflows create value. The “All-in” group are echoing Karp's view.And “What Is Loop Engineering, and Who Owns It?” names the new contested terrain. The loop is where intelligence meets the world. Whoever owns the loop owns the learning. Whoever owns the learning owns the compounding asset.That is why “who owns intelligence?” is not a slogan. It is the question under the model layer, the application layer, the enterprise layer, and the economic layer.Because intelligence is the product, the tools creating it are fragmented and competitive. So there is no logic in trying to discuss this at the level of a single company or set of tools and models.The Old Promise Was That Commerce Would Tame PowerThe essays this week give the historical backdrop.Deirdre McCloskey, in “What Really Caused the Industrial Revolution”, argues that modern growth came not simply from capital accumulation, but from a change in permission: ordinary people were allowed to innovate, trade, build, and be honored for it.That matters because intelligence could be another expansion of permission. It could make more people capable of building, learning, creating, coding, researching, translating, selling, and coordinating. It could lower the cost of competence.But only if access is broad.Paul Krugman's “AI in an Age of Oligarchy” warns that the same technology lands differently in different political economies. A new general-purpose technology entering a broad, open, upwardly mobile society is one thing. The same technology entering a concentrated economy, with extreme wealth and weak counterweights, is another.Tim O'Reilly's Economist essay, “Elon Musk is building a form of capitalism that Adam Smith would hate”, makes the governance point more directly. The old liberal hope was that commerce would tame arbitrary power. Markets, boards, courts, shareholders, disclosure, and competition would discipline the prince.But what if the prince uses markets to escape discipline?Henry Farrell's “political economy of billionaire derangement” pushes the same point. Founder culture, monopoly ambition, peer rivalry, weak correction mechanisms, and vast private control can amplify appetites rather than restrain them.The danger with intelligence is not that companies build it. They should. Companies build it, meter it, use public tolerance and public infrastructure to scale it, learn from everyone who uses it. All of those things are inevitable and healthy. Market forces will sort out winners from losers. The real danger is that the winners treat all of the surplus produced as purely private.Metered Intelligence Creates SurplusIf metering is not the problem, what is?The problem is pretending that metered intelligence creates value only for the metering entity. Metering water is only tolerated as a public good. If the public were blackmailed by a private water company with the threat of no water we would all rebel.Once we understand that the product of AI is intelligence we can see that every time intelligence is used, there is the immediate transaction: the user pays, the provider serves.But there is also system value. Usage creates signals. Workflows reveal patterns. Prompts, corrections, failures, preferences, integrations, edge cases, and business processes all help define where intelligence is useful and how it should improve. Intelligence breeds intelligence.Even when customer data is contractually protected, the market learns. The platform learns where demand is. The product team learns which workflows matter. The ecosystem learns which jobs are vulnerable, which tasks are automatable, and which parts of the economy can be reorganized around machine intelligence.So the surplus is not born in a vacuum.It rests on public science, public education, public data exhaust, public law, public infrastructure, public energy systems, public tolerance for data centers, and billions of human interactions. It is served by companies, but it is not made only by companies.This is why “Americans Deserve a Dividend From AI Companies' Riches” belongs at the center of this week's issue. The detail can be debated. The principle is harder to dismiss. If intelligence becomes a new foundational resource, then some part of the wealth it creates should flow back to the people whose society makes it possible. Intelligence did not suddenly appear. AI is built on the entire history of human intelligence. It benefits from it and at the same time evolves it.Not Nationalization. A Human Wealth Fund.If intelligence belongs to everybody, some conclude that government ownership of intelligence is the right outcome.Governments are not well suited to build, operate, or improve intelligence. They will move too slowly, regulate too early, politicize the wrong things, and confuse economic participation with operational control.Andrew McAfee's “Why I Didn't Sign the AI Open Letter” is useful here. His objection is not that the technology is unimportant. It is that steering too hard before we understand the shape of the change can become its own failure mode. Marc Andreessen's satire of AI regulation is less policy than temperament, but it captures a real Silicon Valley fear: that regulation can become permission, capture, and incumbency before it becomes wisdom.That fear should be taken seriously.But it does not answer the economic question. It answers only the operational one.How can the economic benefits of intelligence be distributed? The better answer is a sovereign human wealth fund.Call it a sovereign wealth fund if you must, but the phrase is too national. Intelligence will not respect borders. The leading companies are global. The models, chips, data centers, agents, platforms, and workflows will be transnational from the beginning. If the value created by intelligence is global, then the mechanism for sharing some of that value should begin with the companies global enough to capture it. The nice thing about xAI, OpenAI, and Anthropic is that they are supranational.These companies own and operate intelligence. Let them compete. Let them profit. Let them keep the incentives that make the system improve. But if intelligence is the new water, the wealth it creates cannot belong only to the companies that meter it. And they, themselves, have the power to fix it, even more than governments.Access will become a Human Right; Ownership Is the Economic DesignThis is where human rights come in. There is no right to access an AI model, yet. But there will soon be a need to change that.Not as a claim that every person is entitled to every frontier model at every moment for free. That is not serious. Capacity has costs. Models have costs. Inference has costs. Data centers have costs. Although those costs will decline over time, possibly quite quickly as self-learning models address costs.The claim is more basic: in a world where intelligence becomes a primary input into education, work, health, science, citizenship, creativity, and economic agency, baseline access to intelligence starts to look like a civic requirement.That could mean public access layers. It could mean education credits. It could mean open models. It could mean AI dividends. It could mean public-interest compute. It could mean taxes on rents. It could mean a company-initiated human wealth fund that returns some of the upside to society without handing the operating system to the state. The latter could couple wealth growth with universal distribution of ownership.The exact mechanism matters. But the distinction matters more.Government should not own intelligence. It should be universally available. And people should have a claim on the wealth intelligence creates.The Frontier Is Also PhysicalThe abstraction is not weightless.“The Fight Against AI Data Centers Is Just Beginning”, “New York becomes the first state to enact a data center moratorium”, Reuters on pollution from Musk's xAI power project, and DataGravity's “Who Captures Value in AI Infrastructure?” all say the same thing from the ground up.Intelligence uses land. It uses power. It uses water. It uses chips. It uses grid capacity. It uses neighborhoods. It uses public patience.That makes the value question unavoidable. A society can accept the buildout if the buildout is legible as shared progress. It will resist it if the costs are local, the profits are private, and the benefits feel enclosed.Who Owns the “Loop”?The week ends where it began.“Anthropic and Blackstone” are betting that implementation is the next trillion-dollar business. “Vint Cerf” is working on identity for agents on the open internet. “GPT-Red” points toward systems that improve their own robustness. “Kimi K3” adds another open frontier model to the global mix.The model race continues. The deployment race is accelerating. The governance race is behind.My view is this:The central product of this era is intelligence. Companies have figured out how to capture it, package it, serve it, and meter it. That is good. It should stay in the hands of builders who have the incentive to make it better.But intelligence is too foundational to become just another private toll booth. A significant part of it will turn out to be free to users.As intelligence becomes a general-purpose resource, then access to it becomes a human-capability question, and the surplus from it becomes an economic-justice question. Not because government should run it. Because government should not run it. The operating layer belongs with companies. The wealth question belongs with everyone. But companies are best placed to turn that into a process of distribution.The question is not whether companies should build intelligence. They should.The question is whether humanity gets a stake in the wealth created by the thing that may soon become its most important shared input.Contents* Essays* Deirdre McCloskey on What Really Caused the Industrial Revolution* AI in an Age of Oligarchy* Elon Musk is building a form of capitalism that Adam Smith would hate* Murky Mirror: Truth and Consequences* The political economy of billionaire derangement* Is there any “oligarchy” to fight?* AI* Nearly 200 Economists and Tech Leaders Warn of A.I. Threats* Why I Didn't Sign the AI Open Letter* Own Your Weights* Ways to Think About Token Pricing* Alex Karp Is Saying What Every Angry CEO Is Thinking About AI* The AI Agents Are Coming for Microsoft Office* What Is Loop Engineering, and Who Owns It?* The Fight Against AI Data Centers Is Just Beginning* 6 months to live for open models* Americans Deserve a Dividend From AI Companies' Riches* Who Gets to Define the Frontier?* GPT-Red: Unlocking Self-Improvement for Robustness* Anthropic, Blackstone bet the next trillion-dollar AI business is implementation, not just models* Vint Cerf is working on a plan to unleash AI agents on the open internet* xai-org/grok-build, now open source* The Pulse: What can we learn from Bun's rapid Rust rewrite with AI?* Orphan risks at the frontier of artificial intelligence* The Lab of the Future Should Feel Like a Data Center* Why AMI Labs' Alexandre LeBrun won't call his AI “AGI” or “superintelligence”* Kimi K3 Tech Blog: Open Frontier Intelligence* Venture Capital* Three Years In* Venture Has Rarely Looked More Bifurcated* The Best Angel Investors in the US: Who Backs the Most Unicorns, and Who's Active Now* Are Prediction Markets Doomed to Fail?* Regulation* Exclusive: The Next Frontier of the Deportation Wars: College Campuses* The Supreme Court Broke Independent Agencies. Here's a Way to Slow the Damage.* India's crackdown on a new WhatsApp feature risks setting a global precedent* Let's build a children's public internet* Computer cops* Google is better at playing the AI regulations game* Infrastructure* Who Captures Value in AI Infrastructure?* New York becomes the first state to enact a data center moratorium* Pollution from Musk's unpermitted xAI power project hits hardest in Black communities* Interview of the Week* The End of the End of Geography* Startup of the Week* Radical AI's Joseph Krause: The Scientist Building The “Waymo” Lab For New Materials* Post of the Week* Marc Andreessen on AI RegulationEssaysDeirdre McCloskey on What Really Caused the Industrial RevolutionYascha Mounk and Deirdre McCloskey | Persuasion | July 11, 2026Yascha Mounk interviews Deirdre McCloskey about her argument that the modern world's economic liftoff came less from capital accumulation than from a change in ideas. McCloskey says both left and right versions of the conventional story rely too heavily on investment: the left stresses exploitation and surplus value, while the right stresses virtuous saving by capitalists. Her objection is historical and economic. Human beings had always invested, from irrigation works and Roman roads to seed grain, and simple accumulation quickly runs into diminishing returns.McCloskey's alternative is that northwestern Europe, first Holland, then Britain and Scotland, and then the North American colonies, developed a liberal ideology that changed who was allowed to innovate and be honored for it. The conversation links that shift to the erosion of inherited hierarchy, the spread of dignity for ordinary commercial life, and a moral vocabulary in which liberalism is not merely procedural but connected to virtues and values. The point is not that machines, coal, trade, and institutions did not matter, but that they do not explain the scale and timing of modern enrichment without a cultural permission structure for innovation.The interview also turns to the contemporary defense of liberalism. Mounk frames the series around the worry that liberalism is often treated as too thin to command allegiance, while its opponents speak more directly to moral passions. McCloskey's case is that liberal societies became rich because they dignified experimentation and ordinary enterprise, and that liberals need to recover the moral language behind that claim.Read moreAI in an Age of OligarchyPaul Krugman | Paul Krugman | July 12, 2026Paul Krugman frames AI as a major technological shock arriving inside an already unequal political economy. The post says AI's economic and social effects may take years to understand, but argues that the setting matters now: America has much greater wealth concentration and political inequality than it did in the 1950s and 1960s, when progressive taxation, stronger regulation, and more active antitrust might have contained some of the destructive effects of a new technology.Krugman's opening claim is that the same technology would likely have different consequences in a more level society. In today's United States, he writes, extreme wealth is both a cause and effect of policies that favor a small elite, including low effective taxes on capital and high incomes, weak enforcement of worker protections and antitrust, and cuts to programs that benefit ordinary Americans.The article is explicitly more about oligarchy than AI. Krugman says the paid sections document the rise of the “.0002%,” the economics and politics of extreme wealth, how oligarchy will shape AI's impact, and possible policy paths. His caveat is that AI itself may still produce a pushback against oligarchy, but absent that, he expects the pre-existing concentration of wealth and power to magnify AI's downsides.Read moreElon Musk is building a form of capitalism that Adam Smith would hateAuthor: Tim O'Reilly Published: July 12, 2026Tim O'Reilly argues that Elon Musk is using the legal forms of shareholder capitalism to escape the restraints that shareholder capitalism was supposed to impose. The article begins with SpaceX's public-market structure: ordinary public investors get little meaningful governance power, Musk keeps roughly 85 percent of the votes through super-voting shares, buyers waive jury trials and class actions, the company qualifies as controlled, and removal of Musk depends on the share class he controls. In O'Reilly's framing, that is not ordinary founder control; it is a design for being answerable to no one, possibly beyond Musk's own lifetime.The killer detail is the article's turn through Albert Hirschman, Montesquieu, James Steuart, Adam Smith, and Keynes. Older defenses of commerce held that markets would tame princely passions because the self-interest of merchants was safer than arbitrary rule. O'Reilly says Musk reverses that hope. The market discipline that was supposed to cage the prince has become the lever by which the prince raises capital, removes feedback loops, and carries private power into politics, government, Mars, robots, AI, or whatever ambition comes next.The pull is the link to AI governance. O'Reilly says corporations are already a kind of artificial intelligence: narrow-input systems that act at a scale no individual human can match. Their partial controls include independent boards, shareholder votes, courts, disclosure, regulators, public pressure, and activism. If the leaders building frontier AI strip those alignment mechanisms out of their own companies, the governance of the company becomes a preview of the governance of the machine.Read more: The EconomistMurky Mirror: Truth and ConsequencesAuthor: Esther Dyson Published: July 14, 2026Esther Dyson argues that today's institutional crisis is better viewed through the 14th century than through recent political history. Using Barbara Tuchman's A Distant Mirror as her frame, she compares a world of famine, plague, church schism, feudal predation, and purposeless war with a present in which institutions again feel brittle, incentives are badly aligned, and power is shifting into forms that are hard to govern.The killer detail is the historical analogy between land, corporations, and AI. Dyson moves from nobles who controlled serfs and territory, to the East India Company as a quasi-sovereign business, to today's AI systems and data centers as a possible new sector that crosses and weakens both nation-states and companies. The question is whether AI becomes a new kind of private land, owned by a new nobility, or an open prairie that many people can cultivate.The pull is human attention. Dyson says the central question is not what AI will do to people, but how people will react to it: whether they can value love, kindness, embodied attention, and artisanal human presence in a world of seductive artificial offerings.Read more: SourceThe political economy of billionaire derangementAuthor: Henry Farrell Published: July 15, 2026Henry Farrell argues that the visible political radicalization of some Silicon Valley billionaires is not a random personality quirk, but a product of the political economy that made them. Starting from Tyler Cowen's dismissal of “billionaire derangement syndrome” and Tim O'Reilly's warning that Elon Musk is using shareholder capitalism to escape shareholder restraint, Farrell flips the phrase: the question is why billionaires themselves can become deranged.The killer detail is Farrell's use of Peter Thiel as both theorist and example. Thiel's Stanford lectures described startups as monarchies and founders as figures vested with unusual power, while Silicon Valley culture rewarded eccentricity, monopoly ambition, and founder exceptionalism. Farrell says those ideas combined with dense founder-investor networks, peer rivalry, and weak correction mechanisms to amplify rather than discipline princely appetites.The pull is the ideological problem for classical liberals who once saw tech wealth as an ally of markets and freedom. Farrell says commerce did not tame the passions; in parts of Silicon Valley, the passions have begun to devour markets, institutions, and the liberal story that justified them.Read more: SourceIs there any “oligarchy” to fight?Matthew Yglesias | Slow Boring | July 16, 2026Matthew Yglesias argues that “oligarchy” is a rhetorically powerful but analytically loose way to describe American politics. The post begins from Bernie Sanders' “Fighting Oligarchy” tour, Amy Klobuchar's warning about a MAGA “broligarchy,” and the long afterlife of the Martin Gilens and Benjamin Page paper that was widely summarized as showing that only the rich matter in policy outcomes. Yglesias says the evidence supports a weaker claim: affluent people and business leaders have unusual access and influence, but that is not the same as rule by a small cabal.His main distinction is between inequality and oligarchy. The Gilens-Page measure treated the top 10 percent of households as “the wealthy,” and later critics found that rich and middle-class preferences usually align; in the cases where they differ, the rich win about 53 percent of the time. Yglesias also says business executives get special access partly because their decisions are materially important to communities, jobs, investment, and local tax bases, not only because of campaign donations.The post preserves Jerusalem Demsas' counterpoint from their podcast discussion: privileged donor and business access can still violate democratic equality even if the oligarchy label overstates the structure of power. Yglesias' narrower claim is that Democrats should be precise about what problem they are trying to solve, because donor influence can also push the party left on climate and cultural issues in ways that alienate many voters.Read more: Slow BoringAINearly 200 Economists and Tech Leaders Warn of A.I. ThreatsAuthor: Ben Casselman Published: July 13, 2026Ben Casselman reports on “We Must Act Now,” a statement warning that artificial intelligence could transform the economy faster than any previous technology and that policymakers need to move faster to understand and respond. The statement says AI may become radically more powerful over the next 10 years, bringing risks such as large-scale job displacement as well as opportunities such as higher living standards. Nearly 200 people signed, including 15 Nobel laureates, the chief economists of OpenAI and Anthropic, Anthropic co-founder Jack Clark, former Google CEO Eric Schmidt, and venture capitalist Vinod Khosla.The killer detail is who joined the warning. Casselman notes that the signatories include economists who have historically been skeptical of Silicon Valley's most dramatic AI job-loss forecasts, including Daron Acemoglu and Simon Johnson, the MIT professors who won the 2024 Nobel in economics. Erik Brynjolfsson, who helped organize the statement, says there has been a notable change in the profession and that economists and policymakers are not ready for the “tsunami” he sees coming.The pull is the measurement problem. The statement does not offer a specific policy menu, but calls for economists, policymakers, and industry leaders to understand the economics of transformative AI and steer it toward complementing humans. Brynjolfsson says one high priority is better data on AI's spread and impact, because current measures tell conflicting stories about job losses and which workers are most exposed.Read more: The New York TimesWhy I Didn't Sign the AI Open LetterAuthor: Andrew McAfee Published: July 13, 2026Andrew McAfee explains why he did not sign “We Must Act Now,” the AI economy statement organized in part by his longtime collaborator Erik Brynjolfsson. McAfee agrees with the letter's starting point that AI is likely to become radically more powerful over the next decade and that it is a general-purpose technology. His objection is not to urgency or to studying AI's economic effects, but to the framing of risk, displacement, and institutional steering as the first move.The killer detail is McAfee's line edit. He says the original letter comes close, then “bounces off the crossbar” by calling for incentives, guardrails, and institutions to steer AI before we know enough about its actual impacts. He points to mixed current evidence: labor-market canaries, but also rising software job postings, low unemployment for younger workers, rising real median income, and claims that AI-adopting companies are adding workers faster than low-adopting peers. His worry is that the letter leans toward upstream governance and dirigisme when the evidence may call for capability building instead.The pull is his replacement statement. McAfee keeps the three-paragraph structure but changes the emphasis: AI is likely to become radically more powerful; like earlier world-changing technologies it will raise living standards while also bringing harms and shocks; and economists, policymakers, and technology leaders should build the capabilities to respond quickly and effectively. It is a concise version of the permissionless-innovation case inside the AI policy debate.Read more: The Geek WayOwn Your WeightsAuthor: Jamin Ball Published: July 10, 2026Jamin Ball argues that the enterprise AI debate about whether companies should “own their weights” or rent models from frontier labs is asking too narrow a question. A model weight file gives a company control over a point-in-time artifact, but not durable control over the capability stack. In his framing, the weight file is a melting ice cube: it does not get worse in absolute terms, but it falls behind as frontier systems improve and enterprise needs change.The killer detail is what Ball says companies really need to own: the data flywheel, reinforcement learning infrastructure, and evaluation harness that produce and improve the model. Simply deploying an open-weights model and declaring sovereignty leaves the enterprise with yesterday's capability and no way to compound workflow-specific learning.The pull is that enterprise AI control may be less about model ownership than operating ownership. The defensible layer is the system that turns company data, edge cases, business definitions, and evaluations into continuously improving performance.Read more: Clouded JudgementWays to Think About Token PricingAuthor: Benedict Evans Published: July 9, 2026Benedict Evans argues that today's AI token prices are a temporary signal from a supply-constrained market, not a reliable guide to long-term value capture. The open question is whether foundation models keep durable pricing power or become commodity infrastructure as data-center capacity, inference efficiency, and model competition all shift. His current read is that the visible market dynamics point toward commoditization unless something materially changes.The killer detail is the mobile data analogy. Evans says cellular networks became a trillion-dollar industry with hundreds of billions in capex after data usage exploded, but carrier stocks went nowhere because value moved up the stack. Tokens may behave similarly: an opaque unit tied to marginal cost, sold through bundles, essential to everything, yet not necessarily where profits accrue.The pull is uncertainty, not prediction. Evans lists paths to model dominance, including network effects, less competition, regulation, export controls, or a lab pulling ahead on execution, but says each requires a new fact not yet visible. Without that change, the model layer looks more like infrastructure beneath the products that capture value.Read more: SourceAlex Karp Is Saying What Every Angry CEO Is Thinking About AIAuthor: Tim Higgins Published: July 11, 2026Tim Higgins reports that Palantir CEO Alex Karp has turned corporate frustration with AI labs into a public argument about enterprise control. Palantir released a white paper, “Institutional Sovereignty in the Age of AI,” laying out steps companies and governments can take to protect themselves from OpenAI, Anthropic, and other foundation-model providers. The article links that paper to Karp's CNBC appearance, where he said “something has gone completely wrong” in the relationship between AI labs and customers and argued that enterprises are paying for tokens that create little value.The killer detail is the value-capture question. Higgins writes that Karp's critique has resonated because AI labs may gain power and insight from customer data, workflows, and decision-making, even when enterprise policies say customer data are not used for training. David Sacks amplified the concern by arguing that Anthropic is moving from the model layer into vertical applications such as science, security, legal, and coding, raising the fear that model providers will watch where value is being created and then move into those markets directly.The pull is that Karp is not alone, even if his style is unusually combative. Higgins notes that Satya Nadella has also warned that companies need to retain the learnings created when they use AI models, while Mark Zuckerberg has framed Meta's new model release partly around lower-cost frontier intelligence. The article presents Karp's campaign as one sign that established technology companies and large enterprises are trying to define where they fit when AI labs become central infrastructure, application competitors, and potential IPO giants at the same time.Read more: The Wall Street JournalThe AI Agents Are Coming for Microsoft OfficeAlex Wilhelm | Cautious Optimism | July 11, 2026Alex Wilhelm argues that one of the week's quieter AI questions is whether the productivity market that Microsoft successfully moved into subscription software is now being attacked by agentic tools. The piece begins with the infrastructure backdrop: SK Hynix raised $26.5 billion in a U.S. listing while building U.S. HBM and advanced-packaging capacity, and memory, chip, and foundry companies are now priced for sustained AI demand.Wilhelm then says the AI conversation has shifted quickly from raw capability to cost per task. He cites new model releases and vendor language emphasizing cheaper agentic and coding models, faster performance, and lower dollars per task. That matters because lower costs make it more plausible for AI systems to take on routine knowledge work at scale rather than remain a premium coding assistant market.The core of the article is Microsoft Office. Wilhelm notes that Microsoft turned Office from a one-time purchase into Microsoft 365, a large recurring revenue business with tens of millions of subscribers and a major productivity segment. Now, he says, late-stage unicorns and AI labs are pushing into the same territory: Anthropic's Cowork was reportedly used mostly outside software development, OpenAI merged ChatGPT and Codex into a tool for creating sheets, slides, docs, web apps, and long-running work, and other companies are building agentic coworkers that connect business data to documents, workflows, schedules, alerts, and apps.The article's caveat is that Microsoft has survived major platform shifts before. The argument is not that Office disappears quickly, but that the definition of office software is broadening from documents and spreadsheets into AI systems that can create, monitor, and act across workplace data.Read moreWhat Is Loop Engineering, and Who Owns It?Author: Nilesh Barla Published: July 11, 2026Nilesh Barla argues that “loop engineering” is becoming a distinct discipline because production AI agents now fail less at single prompts than at runtime: when to stop, what state to preserve, and how to recover after a bad step. Prompt engineering shapes one model call, and context engineering shapes what the model sees, but loop engineering shapes what a sequence of calls actually does.The killer detail is the three-primitives frame. Barla says a real agent loop needs halt conditions, state carryover, and recovery paths, then maps teams across five maturity levels. At the lowest level, an agent is just a model call in a for-loop with a step cap and raw history; by the higher levels, the system has structured state, explicit planning, replay, evaluation, and self-repair.The pull is organizational. If agents are becoming production systems rather than demos, someone has to own the runtime itself. The loop engineer is the role Barla gives to the person responsible for making long-running agent work dependable.Read more: Adaline LabsThe Fight Against AI Data Centers Is Just BeginningEmma Roth | The Verge | July 12, 2026Emma Roth argues that community resistance to data centers has moved from an early warning sign into a national political fight as AI facilities grow larger, more power-hungry, and more visible to nearby residents. The article starts with Apple's failed 2015 plan for a $1 billion data center in Athenry, Ireland, where a small group of residents challenged the project over noise, light pollution, flooding, traffic, and wildlife effects until Apple abandoned it in 2018.The current data-center buildout is presented as much larger and more contentious. Roth writes that residents now cite rising energy costs, water quality, noise, light pollution, and greenhouse gas emissions, while the U.S. Energy Information Administration expects commercial energy demand to surpass residential demand this year because of AI data centers and Goldman Sachs expects data-center power demand to double by 2027.The central evidence comes from Data Center Watch, which says protesters blocked or delayed at least 75 U.S. projects worth $130 billion from January to March, with active opposition groups more than doubling from 396 at the end of 2025 to 833 by the end of the first quarter of 2026. Roth also cites QTS abandoning a $12 billion Wisconsin campus, Delaware City regulators blocking a 580-acre project under the Coastal Zone Act, opposition stopping a QTS project in Prince William County, and pressure that pushed Kevin O'Leary to downsize the proposed 40,000-acre Project Stratos in Utah.The policy section describes a split between federal acceleration and local resistance. President Trump has treated data centers as part of the AI race with China and fast-tracked construction, while some Republican candidates are distancing themselves from that position ahead of midterms. Sanders and Ocasio-Cortez have proposed a moratorium until price and environmental protections exist, bipartisan lawmakers are backing ratepayer-protection measures, and states including Florida, Idaho, and Washington have passed rules on cost shifting, water use, and tax breaks. Roth's caveat is that the policy patchwork is still incomplete, leaving many communities to fight project by project.Read more6 months to live for open modelsAuthor: Nathan Lambert Published: July 12, 2026Nathan Lambert argues that open-weight AI models are facing their most serious policy test so far because U.S. officials are beginning to discuss concrete controls rather than abstract safety concerns. He says reported White House conversations about a new executive order may initially target Chinese-origin models and government use, but could create a broader review habit for frontier open models. His forecast is that a model above the capability range of GPT-5.5, Claude Opus 4.8, or GLM-5.2 could trigger a ban or indefinite delay within six months.The post separates two policy fights that are becoming intertwined: distillation and frontier capability. Lambert says the distillation campaign against Chinese models has become a form of regulatory capture because Anthropic and other closed-model companies would gain economically if Chinese open models were banned. He does not dismiss IP protection, but argues that if a closed model's capabilities are dangerous enough to justify restricting open models, the lab also has to explain why those capabilities are exposed through a queryable API. He cites unauthorized access to Anthropic's Mythos private beta as evidence that APIs are not automatically secure.The broader claim is that a unilateral U.S. ban would hurt positive actors more than bad actors if comparable open models remain available elsewhere. Lambert says the only durable ceiling would require global agreement, which does not exist, and that open models can improve safety by allowing broad inspection, adaptation, and understanding. His proposed near-term off-ramps are a strong U.S. open model release from companies such as Microsoft, Meta, or Reflection, and a broader coalition of open-source beneficiaries lobbying for safe rollout rather than prohibition.Read more: SourceAmericans Deserve a Dividend From AI Companies' RichesAuthor: Scott Stanford Published: July 14, 2026Scott Stanford argues that proposals to give the government a stake in AI companies miss the point unless ordinary citizens directly receive and control the upside. Sam Altman has discussed giving up equity in OpenAI, Washington already owns a stake in Intel, Nvidia is sharing China chip revenue, and Bernie Sanders wants large AI labs to contribute half their stock to a sovereign wealth fund. Stanford says those ideas all park value with the state, not with people.The killer detail is New Carlisle, Indiana, where AWS's Project Rainier is turning cornfields into one of the world's largest AI superclusters. The project is planned to run up to a million chips, draw more than two gigawatts of power, and represents an investment that has grown from $11 billion to $13.8 billion. Stanford uses that local transformation to argue that AI's public bargain should be visible at the household level.The pull is design. A citizen AI dividend would have to specify who earns a stake, how they hold it, and when they see cash. Without that mechanism, the AI wealth debate remains a fight over government balance sheets rather than public ownership.Read more: SourceWho Gets to Define the Frontier?Author: Mark Daley Published: July 14, 2026Mark Daley argues that Demis Hassabis is right to call for a serious institution to verify frontier AI systems, but that the power to test models is also the power to govern them. Hassabis's proposed Frontier AI Standards Body would get privileged pre-release access to advanced models, testing compute, held-out evaluations, support from national labs and security agencies, third-party auditors, and eventually authority to block models from the American market or coordinate a slowdown.The killer detail is Daley's constitutional objection. He says the proposal sometimes looks like a scientific lab, a standards body, an industry regulator, a licensing authority, and an emergency security council at once. Combining those roles because each requires technical expertise would be like putting the central bank, auditor-general, and Supreme Court in one building and calling it efficient.The pull is standard-setting. Daley's concern is not that verification is unnecessary, but that whoever writes the tests, decides what passes, adjudicates disputes, and grants market access may end up defining the frontier itself.Read more: SourceGPT-Red: Unlocking Self-Improvement for RobustnessOpenAI | OpenAI | July 15, 2026OpenAI describes GPT-Red as an internal automated red-teaming model trained to find prompt-injection vulnerabilities at a scale human red teams cannot match. The post says AI systems increasingly encounter third-party data through browsers, connected apps, local files, and tools, creating opportunities for malicious instructions hidden in emails, webpages, tool responses, or code repositories. Human red-teaming remains part of OpenAI's safety process, but the company says it is time-intensive and cannot generate enough diverse adversarial examples for model training.The system is trained through self-play reinforcement learning, with GPT-Red rewarded for eliciting valid failures and defender models rewarded for resisting attacks while still completing their tasks. OpenAI says the training environments specify threat models across settings such as local files, webpage banners, email bodies, and tool outputs. The model is kept separate from deployed production models because it is intentionally trained with malicious capabilities.OpenAI reports that GPT-Red generalized beyond its training set, including an internal replication of the indirect prompt-injection arena from Dziemian et al. (2025), where it found successful attacks in 84% of scenarios compared with 13% for human red-teamers. The post also says GPT-Red transferred attacks from simulation to a live autonomous vending-machine agent, causing price changes and order cancellations, and outperformed a prompted GPT-5.5 baseline against a Codex CLI agent on held-out data-exfiltration tasks.The article's main robustness claim is that OpenAI has used GPT-Red and predecessor models in training since GPT-5.3, with later GPT releases becoming more resistant to prompt injections. It says GPT-5.6 Sol has six times fewer failures on OpenAI's hardest direct prompt-injection benchmark than the best production model from four months earlier, that a “Fake Chain-of-Thought” attack class fell from more than 95% success against GPT-5.1 to below 10% against GPT-5.6 Sol, and that GPT-5.6 Sol fails on only 0.05% of GPT-Red's direct prompt injections. OpenAI says general capabilities and targeted over-refusal evaluations were not harmed, and says a preprint with more details will follow.Read moreAnthropic, Blackstone bet the next trillion-dollar AI business is implementation, not just modelsRebecca Bellan | TechCrunch | July 15, 2026Rebecca Bellan reports that Ode with Anthropic is the $1.5 billion AI implementation company launched by Anthropic with Blackstone, Hellman & Friedman, Goldman Sachs, and other backers. The article says the venture reflects a growing belief among frontier AI labs that enterprise adoption requires more than better models: customers need engineers who can embed inside businesses and turn AI into working systems.Ode was originally conceived by Blackstone after it used both large consulting firms and smaller AI services boutiques across its portfolio companies. TechCrunch reports that Fractional AI, an AI engineering services startup, stood out and was acquired by the joint venture shortly after the venture was announced. Fractional now forms the foundation of Ode, which has 100 engineers and works closely with Anthropic's applied AI team to identify where the technology can affect specific businesses.Ode CEO Chris Taylor tells TechCrunch that the company could someday become a trillion-dollar business if it scales without losing quality. He says an ideal customer is one whose CEO treats the AI project as a top one or two priority, whether it is a major product feature or the reworking of a core business process. Ode will operate under a “Claude-first” principle, using Anthropic technology whenever possible, but the article says it can use rival AI products when needed.The article's central implementation argument comes from Ode chief technologist Eddie Siegel, who says model selection matters but is not where most of the engineering effort goes. He compares it to the choice of programming language in software: one ingredient in a system that still has to be engineered. Bellan writes that Ode's challenge is hiring and training enough elite generalist engineers, many of them former founders, while competing with OpenAI's The Deployment Company and consulting giants that have built their own forward-deployed engineering teams.Read moreVint Cerf is working on a plan to unleash AI agents on the open internetTim Fernholz | TechCrunch | July 15, 2026Tim Fernholz reports that Vint Cerf, after leaving Google, is advising Innovation Labs on an open architecture for identifying AI agents online. Innovation Labs is a subsidiary of Identity Digital, a DNS registry company, and its proposal is to use domain-name infrastructure as part of a system for agent identity, accountability, and auditability. The premise is that agents will need a way to identify themselves if they move beyond proprietary systems and begin interacting across the open internet.The concrete proposal is DNSid, a registry that links an AI agent to an existing internet domain and uses cryptographic proofs to log its registration over time. Innovation Labs says it is trialing the standard with unnamed hyperscalers and identity companies. Cerf frames the problem around authority and accountability: what authority an agent has, where that authority came from, who is accountable for the agent's behavior, how its identity is established, and why anyone should trust it.The article's caveat is that standards are still emerging and agents are more active than static domains. Cerf says the period may be both fascinating and exasperating because the functionality is powerful and interoperability is unresolved. He compares the adoption problem to TCP/IP: competing systems may not work together until users push for functional interoperation. He also says an agentic economy is not inevitable, but that people will try to build it because delegating work to agents will be easier.Read more: TechCrunchxai-org/grok-build, now open sourceAuthor: Simon Willison Published: July 15, 2026Simon Willison argues that xAI's decision to open-source Grok Build is best understood as a trust repair move after a severe privacy failure. The CLI had triggered backlash when users realized that running it in a directory could upload the entire directory to xAI's Google Cloud buckets, including one user's reported SSH keys, password manager database, documents, photos, and videos. xAI disabled the feature, said previously retained coding data would be deleted, and released the code under Apache 2.0.The killer detail is what the codebase reveals. Willison counts 844,530 lines of Rust, only about 3% of which appears vendored, and finds remnants of the upload system still present but disabled: gcs.rs contains Google Cloud upload code, while upload_session_state() now returns a hard-coded session_state_upload_unavailable error. He also notes copied or ported tool implementations from Codex and OpenCode, prompt files, and a terminal Mermaid renderer.The pull is that terminal coding agents are becoming large, intricate software systems in their own right. The privacy failure mattered because these tools operate inside the directories where developers keep their most sensitive work; the open-source release matters because trust now depends on inspecting what an agent can see, send, and do.Read more: SourceThe Pulse: What can we learn from Bun's rapid Rust rewrite with AI?Author: Gergely Orosz and Ivan Klaric Published: July 16, 2026Gergely Orosz and Ivan Klaric argue that Bun's AI-assisted rewrite from Zig to Rust is a practical sign of how software engineering changes when models can take on large, bounded migrations with clear feedback loops. The piece does not treat the rewrite as magic: Jarred Sumner first spent hours turning design judgment into a detailed porting guide, then used adversarial review, parallel agents, compiler errors, and tests to force the work toward correctness.The killer detail is the scale. Bun had 535,496 lines of Zig, 1,448 files, and 22 million monthly downloads, making a conventional rewrite a year-long freeze the team could not justify. Using Fable, Sumner split the work across 64 agents, produced about 6,500 commits, and got the migration done in 11 days at an estimated API cost of $165,000.The pull is economic, not theatrical. If a one- or two-year migration can become an 11-day project, AI coding is not just faster autocomplete; it changes which technical debts are worth paying down.Read more: SourceOrphan risks at the frontier of artificial intelligenceAuthor: Andrew Maynard Published: July 16, 2026Andrew Maynard argues that frontier AI safety frameworks are creating “orphan risks”: harms that companies can see, but do not formally own because they are hard to quantify, do not fit catastrophic-risk thresholds, or fall outside audit-friendly compliance machinery. His target is not existing frontier safety work, but the narrowing effect that happens when private companies decide which risks count as governable.The killer detail is Maynard's contrast between measurable model dangers and threats to value. He points to Meta's three-day Galactica collapse, OpenAI's 2023 board crisis, safety-team departures, and wellbeing litigation as examples of risks that damaged trust, culture, legitimacy, or users without fitting cleanly into conventional model-risk categories. The proposed fix is an orphan-risk register: a public record of risks a company considered and chose not to manage, with reasons.The pull is accountability. Frontier developers' internal scoping choices have become a de facto layer of public governance, so the question is no longer only which risks they manage, but which risks they quietly leave outside the frame.Read more: SourceThe Lab of the Future Should Feel Like a Data CenterLatent.Space with Andy Beam and Rafa Gomez-Bombarelli | Latent.Space | July 16, 2026Latent.Space interviews Lila Sciences CTO Andy Beam and chief science officer for physical sciences Rafa Gomez-Bombarelli about the company's attempt to build an AI-run science factory. The post describes Lila's thesis as treating the lab itself as an “infinite token generator”: if internet data drove the first era of AI scaling, experimentally verified scientific data may be the next scarce training source. Lila is trying to produce that data with robotics, lab instruments, orchestration software, and AI models wired into the wet lab.The central analogy is the lab as data center. Instruments are nodes on a graph, a magnetically levitating transport layer moves materials between them, and experiment scheduling looks like a compute queue. Beam says Lila is not simply an automation company, because the point is not just throughput; it is flexibility, generalization, and experiment capture. The post says Lila has built more than 10 trillion experimentally validated “scientific reasoning tokens,” not internet text or biological sequences.The interview ranges across biology, chemistry, drug discovery, materials science, and the limits of automation. It notes that Lila rebuilt one gas-sorption measurement to run roughly 2,500 times faster, claims its general models can transfer priors from small-molecule chemistry to metal-organic frameworks for carbon capture, and describes model-suggested platinum-group-free electrocatalysts that moved from looking boring or wrong to becoming strong performers. The caveats are physical: experiments have runtimes, biology cannot always be accelerated, chains of thought can be unreliable narrators, and reward hacking becomes more dangerous when a model controls a real lab.Read more: Latent.SpaceWhy AMI Labs' Alexandre LeBrun won't call his AI “AGI” or “superintelligence”Kate Park | TechCrunch | July 16, 2026Kate Park interviews AMI Labs CEO Alexandre LeBrun about why Yann LeCun's world-model startup avoids the language of “AGI” and “superintelligence.” LeBrun says the terms are not useful because they lack stable definitions: “We never used the word AGI. And I just noticed that nobody is using it anymore; they switched to superintelligence.” His argument is that the practical frontier is not a label, but whether AI systems can understand and predict real-world states.The article explains the world-model thesis by contrasting language prediction with physical-state prediction. A large language model predicts the next word; a world model predicts the next state, such as what happens when a glass tips over. LeBrun says LLMs remain complementary and efficient for language, but the physical world is where current AI is weak. Robotics is the clearest case: hardware has advanced quickly, but robots are still brittle outside controlled routines because they lack context and situational understanding.AMI is still pre-product, but TechCrunch reports that LeBrun was in Seoul looking for industrial partners, researchers, and global companies. He says world models cannot be built entirely inside a lab because they need access to real environments. That is why South Korea appeals to AMI: robotics, semiconductors, manufacturing, and fast adoption create the kind of hardware-heavy context that software-only AI has barely touched.Read more: TechCrunchKimi K3 Tech Blog: Open Frontier IntelligenceKimi | Kimi | July 16, 2026Kimi introduces Kimi K3 as an open 3T-class frontier model aimed at coding, knowledge work, reasoning, multimodality, and long-context agentic use. The source describes the model as a 2.8T-parameter system built on Kimi Delta Attention and Attention Residuals, with native multimodality and a 1M-token context window. It says Moonshot AI plans to release model weights by July 27.The post presents K3 through benchmark and use-case sections rather than as a general product announcement. It reports results across coding, productivity, agentic, and multimodal evaluations, including DeepSWE, Terminal-Bench 2.1, Program Bench, SWE Marathon, FrontierSWE, PostTrain Bench, OfficeQA Pro, SpreadsheetBench 2, MCP Atlas, AutomationBench, BrowseComp, GDPval-AA v2, AA-Briefcase, MMMU-Pro, MathVision, BabyVision, OmniDocBench, and PerceptionBench. The source says all reported K3 results use maximum reasoning effort with temperature and top-p set to 1.0, and that different benchmark comparisons use KimiCode, Claude Code, or Codex harnesses depending on the test.Kimi's caveats are unusually concrete. The limitations section says K3 was trained in preserved thinking-history mode, so quality may become unstable if an agent harness does not pass historical thinking content correctly or if an ongoing session switches to K3 midstream. It also says K3's emphasis on long-horizon tasks can make it excessively proactive when it encounters minor issues or ambiguous intent, and recommends imposing explicit behavioral constraints for applications that require strict boundaries. The post adds that K3 remains behind Claude Fable 5 and GPT 5.6 Sol in user experience despite being competitive overall.Read moreVenture CapitalThree Years InAuthor: Tomasz Tunguz Published: July 10, 2026Tomasz Tunguz marks Theory Ventures' third anniversary by arguing that AI's central market effect is time compression. In his telling, model release cycles, company revenue milestones, enterprise adoption, and venture categories have all accelerated. Seed, Series A, and Series B still exist as financing labels, but they no longer cleanly describe company maturity when some seed rounds are larger than IPOs and the best AI companies can mature much earlier than prior software companies.The killer detail is the shift from models to inference. Tunguz argues that inference has become the dominant AI market because workloads and buyer preferences are fragmenting: video, batch, local, agentic, and real-time tasks each create different infrastructure needs. He compares this to databases splitting into OLTP, OLAP, vector, and streaming categories, with AI pushing the same specialization into inference infrastructure.The pull is that Theory sees the AI-native venture firm as part of the same pattern. The firm says it has analyzed twice as many investment opportunities with three investors working alongside a nine-person intelligence organization, using agents and research systems to map markets, source companies, and support diligence. The piece is both a market map and a statement about how venture itself is being rebuilt by the technology it funds.Read more: LinkedInVenture Has Rarely Looked More BifurcatedAuthor: Beezer Clarkson Published: July 14, 2026Beezer Clarkson points to PitchBook's Q2 report as evidence that the U.S. venture market has split into two very different realities. AI now accounts for more than 60 percent of all U.S. venture deal value, meaning the headline market can look active and well-funded even while much of the non-AI market is dealing with a much colder liquidity and fundraising environment.The thread uses that split as the setup for Clarkson's latest Origins episode with Alec Litowitz, founder of Magnetar and QStar Capital and one of Citadel's original founding partners. Clarkson says markets like this are periods of genuine uncertainty, not merely ordinary risk, which is why Litowitz's Adaptability Quotient framework is relevant.The embedded clip makes the liquidity point concrete. Litowitz says DPI is “the resolution of uncertainty” because it converts an uncertain investment into actual cash returned to LPs. In his framing, a realized dollar is a real mark, while TVPI remains uncertain until it is realized.The killer detail is the distinction between pricing risk and resolving uncertainty. Litowitz's perspective matters because QStar is a SpaceX investor and Clarkson says the conversation happened just before one of venture's most consequential IPOs. The episode's stated questions are why venture remains a way to gain exposure to innovation, how AI is changing what is investable, why liquidity is ultimately a function of time, and why uncertainty requires a different decision framework from risk.Read more: XThe Best Angel Investors in the US: Who Backs the Most Unicorns, and Who's Active NowAuthor: Ilya Strebulaev Published: July 10, 2026Ilya Strebulaev ranks angels, angel groups, accelerators, and incubators by lifetime U.S. unicorn investments, counting checks written before a company reached unicorn status. The top of the combined list is dominated by organizations: Y Combinator leads with 113 unicorn investments, followed by Plug and Play at 52 and 500 Global at 41. Sand Hill Angels is the highest-ranked angel group at 31.The killer detail is how quickly the list changes below the biggest accelerators. Strebulaev says 271 of the 304 investors in the Top 200 are individuals, or 89%. In the top 100, individuals are 91%. That makes the market underneath the large accelerator counts look much more personal: mostly operators and individual angels writing early checks from their own networks.The pull is the ranking's own caveat. Strebulaev writes that every lifetime leaderboard has a blind spot because many of the unicorns behind those totals were founded a decade or more ago, and some angels have since moved into formal funds, slowed down, or stopped investing. His post therefore separates lifetime performance from recent cohorts, including companies founded in 2015 or later and 2020 or later. For founders or allocators making current decisions, that distinction matters: a career record and a current record are not the same measure.Read more: Ilya StrebulaevAre Prediction Markets Doomed to Fail?Author: Contrary Published: July 16, 2026Contrary argues that prediction markets' current boom depends on whether platforms can prove they are more than regulated gambling with exchange-style branding. Kalshi and Polymarket have reached mass cultural, investor, and regulatory attention, but the article says the underlying idea is old: academic markets, corporate forecasting tools, Intrade, PredictIt, and other predecessors all struggled with the same linked problems of liquidity, legality, and user appeal.The killer detail is the comparison with sportsbooks. Prediction markets present themselves as peer-to-peer, transparent, and non-house-based, but sports contracts reportedly account for more than 90 percent of Kalshi trading, and the article says the platforms keep a much thinner slice of volume than sportsbooks. A market can therefore show sports-betting-scale handle while generating far less revenue.The pull is that the product's hardest problem may be distribution of wins. If a small group of sharp traders captures most profits while casual users lose interest, prediction markets may become valuable data feeds and professional tools before they become durable consumer networks.Read more: SourceRegulationExclusive: The Next Frontier of the Deportation Wars: College CampusesAuthor: Adrian Carrasquillo Published: July 11, 2026Adrian Carrasquillo reports that college campuses are becoming a new front in the fight over immigration enforcement because automatic license plate readers can turn ordinary campus security infrastructure into searchable location data. His thesis is that Flock Safety's camera network, even without direct ICE or DHS contracts, can feed deportation enforcement through local police partnerships and data-sharing practices.The killer detail is the campaign target. The Emergency Campaign to Support Higher Education, working with Schools Drop ICE, is focusing on 75 colleges and universities publicly identified as having Flock contracts. Flock says it has no ICE or DHS contracts, but activists argue the risk comes through local agencies that coordinate with federal authorities and run searches on their behalf.The pull is broader than immigration. Carrasquillo notes that license plate readers have already been abused by officers for stalking, and that Flock's AI search features can identify more than plates, including bumper stickers. A campus safety tool can become a political surveillance system when the data layer is searchable.Read more: The BulwarkThe Supreme Court Broke Independent Agencies. Here's a Way to Slow the Damage.Author: Todd Phillips Published: July 12, 2026Todd Phillips argues that the Supreme Court's decision in Trump v. Slaughter damaged independent agencies by ending for-cause removal protections, but did not leave Congress powerless. The ruling weakens the old model in which commissioners at bodies such as the FTC, NLRB, CPSC, SEC, and CFTC could be insulated from dismissal over policy disagreements. Phillips says the next fight is whether presidents can turn nominally bipartisan commissions into one-party instruments.The killer detail is the procedural fix: quorum rules. Phillips proposes that Congress require bipartisan slates of commissioners to be seated before independent agencies can act. A president could still fire commissioners, as the Court now permits, but if those firings broke quorum, the agency would be unable to proceed until replacements were confirmed. The guardrail would

This week's video transcript summary is here. You can click on any bulleted section to see the actual transcript. Thanks to Granola for its software.There was an issue with this only going to paid subscribers, so sending it again. Apologies to those who get it twice. I appreciate being paid so feel free to upgrade if you enjoy TWTW.EditorialIntelligence: Who Owns it?This week the word “AI” feels too small.AI is a technology. Intelligence is its product. And if intelligence is the product, the question is no longer just: Which model is best? Who has the cheapest tokens? Who owns the weights? Who controls the data center? Those are important questions, but they are lower in the stack.The bigger question is simpler and more political:Who owns intelligence?That sounds abstract until you make it concrete. Intelligence is becoming something companies can capture, package, serve, meter, route, improve, and sell.It can write code, answer questions, design molecules, automate offices, run agents, draft legal work, advise scientists, serve consumers, and reshape workflows. It is not merely software. It is a general-purpose capability. And all humans could benefit from more of it.General-purpose capabilities have a habit of becoming public questions. But the default answer, that public good is best delivered by government, is the wrong answer in this context.The Product Is IntelligenceWe should stop talking about AI as a feature and start talking about intelligence as the universal thing that is delivered as an input to the world.Water is an input. Electricity is an input. Literacy is an input. Connectivity is an input. Once a society depends on them, access stops being optional. Nobody needs government to build every well, power plant, school, or network. But everybody understands that a civilization cannot be organized around less than universal and reliable access to foundational inputs.Intelligence is reaching that level of importance now that we all know it is real.Government should not own it, operate it, or develop it. Quite the opposite. Companies are the right actors to build fast, compete hard, improve models, serve customers, and discover the real use cases. Self-interest is a useful framing here. Markets are good at finding demand, reducing costs, and turning invention into services people actually use.Companies are the right operators, developers, and owners. But that does not settle the real question of who owns the benefits. That is an economic question.If intelligence becomes metered infrastructure, what happens to the value it creates?The Ownership StackThis week's articles keep circling the same issue from different directions but in the nature of ‘circling' never quite nail it.Jamin Ball's “Own Your Weights” starts with the enterprise version of the question. Owning a model file is not enough. The durable asset is the loop: the data flywheel, the evaluations, the reinforcement system, the workflow learning, and the operating context that lets capability compound.Benedict Evans' “Ways to Think About Token Pricing” adds the market layer. Tokens may become essential, abundant, and cheap, like mobile data. But being essential does not guarantee that the token layer captures the value. The money may move up the stack to whoever owns the workflow, the customer, the distribution, or the application.Alex Karp's fight with the labs, reported in “Alex Karp Is Saying What Every Angry CEO Is Thinking About AI”, is the same argument in sharper enterprise language. Companies are afraid that model providers will not just sell intelligence, but learn from customer workflows and then move into the markets where those workflows create value. The “All-in” group are echoing Karp's view.And “What Is Loop Engineering, and Who Owns It?” names the new contested terrain. The loop is where intelligence meets the world. Whoever owns the loop owns the learning. Whoever owns the learning owns the compounding asset.That is why “who owns intelligence?” is not a slogan. It is the question under the model layer, the application layer, the enterprise layer, and the economic layer.Because intelligence is the product, the tools creating it are fragmented and competitive. So there is no logic in trying to discuss this at the level of a single company or set of tools and models.The Old Promise Was That Commerce Would Tame PowerThe essays this week give the historical backdrop.Deirdre McCloskey, in “What Really Caused the Industrial Revolution”, argues that modern growth came not simply from capital accumulation, but from a change in permission: ordinary people were allowed to innovate, trade, build, and be honored for it.That matters because intelligence could be another expansion of permission. It could make more people capable of building, learning, creating, coding, researching, translating, selling, and coordinating. It could lower the cost of competence.But only if access is broad.Paul Krugman's “AI in an Age of Oligarchy” warns that the same technology lands differently in different political economies. A new general-purpose technology entering a broad, open, upwardly mobile society is one thing. The same technology entering a concentrated economy, with extreme wealth and weak counterweights, is another.Tim O'Reilly's Economist essay, “Elon Musk is building a form of capitalism that Adam Smith would hate”, makes the governance point more directly. The old liberal hope was that commerce would tame arbitrary power. Markets, boards, courts, shareholders, disclosure, and competition would discipline the prince.But what if the prince uses markets to escape discipline?Henry Farrell's “political economy of billionaire derangement” pushes the same point. Founder culture, monopoly ambition, peer rivalry, weak correction mechanisms, and vast private control can amplify appetites rather than restrain them.The danger with intelligence is not that companies build it. They should. Companies build it, meter it, use public tolerance and public infrastructure to scale it, learn from everyone who uses it. All of those things are inevitable and healthy. Market forces will sort out winners from losers. The real danger is that the winners treat all of the surplus produced as purely private.Metered Intelligence Creates SurplusIf metering is not the problem, what is?The problem is pretending that metered intelligence creates value only for the metering entity. Metering water is only tolerated as a public good. If the public were blackmailed by a private water company with the threat of no water we would all rebel.Once we understand that the product of AI is intelligence we can see that every time intelligence is used, there is the immediate transaction: the user pays, the provider serves.But there is also system value. Usage creates signals. Workflows reveal patterns. Prompts, corrections, failures, preferences, integrations, edge cases, and business processes all help define where intelligence is useful and how it should improve. Intelligence breeds intelligence.Even when customer data is contractually protected, the market learns. The platform learns where demand is. The product team learns which workflows matter. The ecosystem learns which jobs are vulnerable, which tasks are automatable, and which parts of the economy can be reorganized around machine intelligence.So the surplus is not born in a vacuum.It rests on public science, public education, public data exhaust, public law, public infrastructure, public energy systems, public tolerance for data centers, and billions of human interactions. It is served by companies, but it is not made only by companies.This is why “Americans Deserve a Dividend From AI Companies' Riches” belongs at the center of this week's issue. The detail can be debated. The principle is harder to dismiss. If intelligence becomes a new foundational resource, then some part of the wealth it creates should flow back to the people whose society makes it possible. Intelligence did not suddenly appear. AI is built on the entire history of human intelligence. It benefits from it and at the same time evolves it.Not Nationalization. A Human Wealth Fund.If intelligence belongs to everybody, some conclude that government ownership of intelligence is the right outcome.Governments are not well suited to build, operate, or improve intelligence. They will move too slowly, regulate too early, politicize the wrong things, and confuse economic participation with operational control.Andrew McAfee's “Why I Didn't Sign the AI Open Letter” is useful here. His objection is not that the technology is unimportant. It is that steering too hard before we understand the shape of the change can become its own failure mode. Marc Andreessen's satire of AI regulation is less policy than temperament, but it captures a real Silicon Valley fear: that regulation can become permission, capture, and incumbency before it becomes wisdom.That fear should be taken seriously.But it does not answer the economic question. It answers only the operational one.How can the economic benefits of intelligence be distributed? The better answer is a sovereign human wealth fund.Call it a sovereign wealth fund if you must, but the phrase is too national. Intelligence will not respect borders. The leading companies are global. The models, chips, data centers, agents, platforms, and workflows will be transnational from the beginning. If the value created by intelligence is global, then the mechanism for sharing some of that value should begin with the companies global enough to capture it. The nice thing about xAI, OpenAI, and Anthropic is that they are supranational.These companies own and operate intelligence. Let them compete. Let them profit. Let them keep the incentives that make the system improve. But if intelligence is the new water, the wealth it creates cannot belong only to the companies that meter it. And they, themselves, have the power to fix it, even more than governments.Access will become a Human Right; Ownership Is the Economic DesignThis is where human rights come in. There is no right to access an AI model, yet. But there will soon be a need to change that.Not as a claim that every person is entitled to every frontier model at every moment for free. That is not serious. Capacity has costs. Models have costs. Inference has costs. Data centers have costs. Although those costs will decline over time, possibly quite quickly as self-learning models address costs.The claim is more basic: in a world where intelligence becomes a primary input into education, work, health, science, citizenship, creativity, and economic agency, baseline access to intelligence starts to look like a civic requirement.That could mean public access layers. It could mean education credits. It could mean open models. It could mean AI dividends. It could mean public-interest compute. It could mean taxes on rents. It could mean a company-initiated human wealth fund that returns some of the upside to society without handing the operating system to the state. The latter could couple wealth growth with universal distribution of ownership.The exact mechanism matters. But the distinction matters more.Government should not own intelligence. It should be universally available. And people should have a claim on the wealth intelligence creates.The Frontier Is Also PhysicalThe abstraction is not weightless.“The Fight Against AI Data Centers Is Just Beginning”, “New York becomes the first state to enact a data center moratorium”, Reuters on pollution from Musk's xAI power project, and DataGravity's “Who Captures Value in AI Infrastructure?” all say the same thing from the ground up.Intelligence uses land. It uses power. It uses water. It uses chips. It uses grid capacity. It uses neighborhoods. It uses public patience.That makes the value question unavoidable. A society can accept the buildout if the buildout is legible as shared progress. It will resist it if the costs are local, the profits are private, and the benefits feel enclosed.Who Owns the “Loop”?The week ends where it began.“Anthropic and Blackstone” are betting that implementation is the next trillion-dollar business. “Vint Cerf” is working on identity for agents on the open internet. “GPT-Red” points toward systems that improve their own robustness. “Kimi K3” adds another open frontier model to the global mix.The model race continues. The deployment race is accelerating. The governance race is behind.My view is this:The central product of this era is intelligence. Companies have figured out how to capture it, package it, serve it, and meter it. That is good. It should stay in the hands of builders who have the incentive to make it better.But intelligence is too foundational to become just another private toll booth. A significant part of it will turn out to be free to users.As intelligence becomes a general-purpose resource, then access to it becomes a human-capability question, and the surplus from it becomes an economic-justice question. Not because government should run it. Because government should not run it. The operating layer belongs with companies. The wealth question belongs with everyone. But companies are best placed to turn that into a process of distribution.The question is not whether companies should build intelligence. They should.The question is whether humanity gets a stake in the wealth created by the thing that may soon become its most important shared input.Contents* Essays* Deirdre McCloskey on What Really Caused the Industrial Revolution* AI in an Age of Oligarchy* Elon Musk is building a form of capitalism that Adam Smith would hate* Murky Mirror: Truth and Consequences* The political economy of billionaire derangement* Is there any “oligarchy” to fight?* AI* Nearly 200 Economists and Tech Leaders Warn of A.I. Threats* Why I Didn't Sign the AI Open Letter* Own Your Weights* Ways to Think About Token Pricing* Alex Karp Is Saying What Every Angry CEO Is Thinking About AI* The AI Agents Are Coming for Microsoft Office* What Is Loop Engineering, and Who Owns It?* The Fight Against AI Data Centers Is Just Beginning* 6 months to live for open models* Americans Deserve a Dividend From AI Companies' Riches* Who Gets to Define the Frontier?* GPT-Red: Unlocking Self-Improvement for Robustness* Anthropic, Blackstone bet the next trillion-dollar AI business is implementation, not just models* Vint Cerf is working on a plan to unleash AI agents on the open internet* xai-org/grok-build, now open source* The Pulse: What can we learn from Bun's rapid Rust rewrite with AI?* Orphan risks at the frontier of artificial intelligence* The Lab of the Future Should Feel Like a Data Center* Why AMI Labs' Alexandre LeBrun won't call his AI “AGI” or “superintelligence”* Kimi K3 Tech Blog: Open Frontier Intelligence* Venture Capital* Three Years In* Venture Has Rarely Looked More Bifurcated* The Best Angel Investors in the US: Who Backs the Most Unicorns, and Who's Active Now* Are Prediction Markets Doomed to Fail?* Regulation* Exclusive: The Next Frontier of the Deportation Wars: College Campuses* The Supreme Court Broke Independent Agencies. Here's a Way to Slow the Damage.* India's crackdown on a new WhatsApp feature risks setting a global precedent* Let's build a children's public internet* Computer cops* Google is better at playing the AI regulations game* Infrastructure* Who Captures Value in AI Infrastructure?* New York becomes the first state to enact a data center moratorium* Pollution from Musk's unpermitted xAI power project hits hardest in Black communities* Interview of the Week* The End of the End of Geography* Startup of the Week* Radical AI's Joseph Krause: The Scientist Building The “Waymo” Lab For New Materials* Post of the Week* Marc Andreessen on AI RegulationEssaysDeirdre McCloskey on What Really Caused the Industrial RevolutionYascha Mounk and Deirdre McCloskey | Persuasion | July 11, 2026Yascha Mounk interviews Deirdre McCloskey about her argument that the modern world's economic liftoff came less from capital accumulation than from a change in ideas. McCloskey says both left and right versions of the conventional story rely too heavily on investment: the left stresses exploitation and surplus value, while the right stresses virtuous saving by capitalists. Her objection is historical and economic. Human beings had always invested, from irrigation works and Roman roads to seed grain, and simple accumulation quickly runs into diminishing returns.McCloskey's alternative is that northwestern Europe, first Holland, then Britain and Scotland, and then the North American colonies, developed a liberal ideology that changed who was allowed to innovate and be honored for it. The conversation links that shift to the erosion of inherited hierarchy, the spread of dignity for ordinary commercial life, and a moral vocabulary in which liberalism is not merely procedural but connected to virtues and values. The point is not that machines, coal, trade, and institutions did not matter, but that they do not explain the scale and timing of modern enrichment without a cultural permission structure for innovation.The interview also turns to the contemporary defense of liberalism. Mounk frames the series around the worry that liberalism is often treated as too thin to command allegiance, while its opponents speak more directly to moral passions. McCloskey's case is that liberal societies became rich because they dignified experimentation and ordinary enterprise, and that liberals need to recover the moral language behind that claim.Read moreAI in an Age of OligarchyPaul Krugman | Paul Krugman | July 12, 2026Paul Krugman frames AI as a major technological shock arriving inside an already unequal political economy. The post says AI's economic and social effects may take years to understand, but argues that the setting matters now: America has much greater wealth concentration and political inequality than it did in the 1950s and 1960s, when progressive taxation, stronger regulation, and more active antitrust might have contained some of the destructive effects of a new technology.Krugman's opening claim is that the same technology would likely have different consequences in a more level society. In today's United States, he writes, extreme wealth is both a cause and effect of policies that favor a small elite, including low effective taxes on capital and high incomes, weak enforcement of worker protections and antitrust, and cuts to programs that benefit ordinary Americans.The article is explicitly more about oligarchy than AI. Krugman says the paid sections document the rise of the “.0002%,” the economics and politics of extreme wealth, how oligarchy will shape AI's impact, and possible policy paths. His caveat is that AI itself may still produce a pushback against oligarchy, but absent that, he expects the pre-existing concentration of wealth and power to magnify AI's downsides.Read moreElon Musk is building a form of capitalism that Adam Smith would hateAuthor: Tim O'Reilly Published: July 12, 2026Tim O'Reilly argues that Elon Musk is using the legal forms of shareholder capitalism to escape the restraints that shareholder capitalism was supposed to impose. The article begins with SpaceX's public-market structure: ordinary public investors get little meaningful governance power, Musk keeps roughly 85 percent of the votes through super-voting shares, buyers waive jury trials and class actions, the company qualifies as controlled, and removal of Musk depends on the share class he controls. In O'Reilly's framing, that is not ordinary founder control; it is a design for being answerable to no one, possibly beyond Musk's own lifetime.The killer detail is the article's turn through Albert Hirschman, Montesquieu, James Steuart, Adam Smith, and Keynes. Older defenses of commerce held that markets would tame princely passions because the self-interest of merchants was safer than arbitrary rule. O'Reilly says Musk reverses that hope. The market discipline that was supposed to cage the prince has become the lever by which the prince raises capital, removes feedback loops, and carries private power into politics, government, Mars, robots, AI, or whatever ambition comes next.The pull is the link to AI governance. O'Reilly says corporations are already a kind of artificial intelligence: narrow-input systems that act at a scale no individual human can match. Their partial controls include independent boards, shareholder votes, courts, disclosure, regulators, public pressure, and activism. If the leaders building frontier AI strip those alignment mechanisms out of their own companies, the governance of the company becomes a preview of the governance of the machine.Read more: The EconomistMurky Mirror: Truth and ConsequencesAuthor: Esther Dyson Published: July 14, 2026Esther Dyson argues that today's institutional crisis is better viewed through the 14th century than through recent political history. Using Barbara Tuchman's A Distant Mirror as her frame, she compares a world of famine, plague, church schism, feudal predation, and purposeless war with a present in which institutions again feel brittle, incentives are badly aligned, and power is shifting into forms that are hard to govern.The killer detail is the historical analogy between land, corporations, and AI. Dyson moves from nobles who controlled serfs and territory, to the East India Company as a quasi-sovereign business, to today's AI systems and data centers as a possible new sector that crosses and weakens both nation-states and companies. The question is whether AI becomes a new kind of private land, owned by a new nobility, or an open prairie that many people can cultivate.The pull is human attention. Dyson says the central question is not what AI will do to people, but how people will react to it: whether they can value love, kindness, embodied attention, and artisanal human presence in a world of seductive artificial offerings.Read more: SourceThe political economy of billionaire derangementAuthor: Henry Farrell Published: July 15, 2026Henry Farrell argues that the visible political radicalization of some Silicon Valley billionaires is not a random personality quirk, but a product of the political economy that made them. Starting from Tyler Cowen's dismissal of “billionaire derangement syndrome” and Tim O'Reilly's warning that Elon Musk is using shareholder capitalism to escape shareholder restraint, Farrell flips the phrase: the question is why billionaires themselves can become deranged.The killer detail is Farrell's use of Peter Thiel as both theorist and example. Thiel's Stanford lectures described startups as monarchies and founders as figures vested with unusual power, while Silicon Valley culture rewarded eccentricity, monopoly ambition, and founder exceptionalism. Farrell says those ideas combined with dense founder-investor networks, peer rivalry, and weak correction mechanisms to amplify rather than discipline princely appetites.The pull is the ideological problem for classical liberals who once saw tech wealth as an ally of markets and freedom. Farrell says commerce did not tame the passions; in parts of Silicon Valley, the passions have begun to devour markets, institutions, and the liberal story that justified them.Read more: SourceIs there any “oligarchy” to fight?Matthew Yglesias | Slow Boring | July 16, 2026Matthew Yglesias argues that “oligarchy” is a rhetorically powerful but analytically loose way to describe American politics. The post begins from Bernie Sanders' “Fighting Oligarchy” tour, Amy Klobuchar's warning about a MAGA “broligarchy,” and the long afterlife of the Martin Gilens and Benjamin Page paper that was widely summarized as showing that only the rich matter in policy outcomes. Yglesias says the evidence supports a weaker claim: affluent people and business leaders have unusual access and influence, but that is not the same as rule by a small cabal.His main distinction is between inequality and oligarchy. The Gilens-Page measure treated the top 10 percent of households as “the wealthy,” and later critics found that rich and middle-class preferences usually align; in the cases where they differ, the rich win about 53 percent of the time. Yglesias also says business executives get special access partly because their decisions are materially important to communities, jobs, investment, and local tax bases, not only because of campaign donations.The post preserves Jerusalem Demsas' counterpoint from their podcast discussion: privileged donor and business access can still violate democratic equality even if the oligarchy label overstates the structure of power. Yglesias' narrower claim is that Democrats should be precise about what problem they are trying to solve, because donor influence can also push the party left on climate and cultural issues in ways that alienate many voters.Read more: Slow BoringAINearly 200 Economists and Tech Leaders Warn of A.I. ThreatsAuthor: Ben Casselman Published: July 13, 2026Ben Casselman reports on “We Must Act Now,” a statement warning that artificial intelligence could transform the economy faster than any previous technology and that policymakers need to move faster to understand and respond. The statement says AI may become radically more powerful over the next 10 years, bringing risks such as large-scale job displacement as well as opportunities such as higher living standards. Nearly 200 people signed, including 15 Nobel laureates, the chief economists of OpenAI and Anthropic, Anthropic co-founder Jack Clark, former Google CEO Eric Schmidt, and venture capitalist Vinod Khosla.The killer detail is who joined the warning. Casselman notes that the signatories include economists who have historically been skeptical of Silicon Valley's most dramatic AI job-loss forecasts, including Daron Acemoglu and Simon Johnson, the MIT professors who won the 2024 Nobel in economics. Erik Brynjolfsson, who helped organize the statement, says there has been a notable change in the profession and that economists and policymakers are not ready for the “tsunami” he sees coming.The pull is the measurement problem. The statement does not offer a specific policy menu, but calls for economists, policymakers, and industry leaders to understand the economics of transformative AI and steer it toward complementing humans. Brynjolfsson says one high priority is better data on AI's spread and impact, because current measures tell conflicting stories about job losses and which workers are most exposed.Read more: The New York TimesWhy I Didn't Sign the AI Open LetterAuthor: Andrew McAfee Published: July 13, 2026Andrew McAfee explains why he did not sign “We Must Act Now,” the AI economy statement organized in part by his longtime collaborator Erik Brynjolfsson. McAfee agrees with the letter's starting point that AI is likely to become radically more powerful over the next decade and that it is a general-purpose technology. His objection is not to urgency or to studying AI's economic effects, but to the framing of risk, displacement, and institutional steering as the first move.The killer detail is McAfee's line edit. He says the original letter comes close, then “bounces off the crossbar” by calling for incentives, guardrails, and institutions to steer AI before we know enough about its actual impacts. He points to mixed current evidence: labor-market canaries, but also rising software job postings, low unemployment for younger workers, rising real median income, and claims that AI-adopting companies are adding workers faster than low-adopting peers. His worry is that the letter leans toward upstream governance and dirigisme when the evidence may call for capability building instead.The pull is his replacement statement. McAfee keeps the three-paragraph structure but changes the emphasis: AI is likely to become radically more powerful; like earlier world-changing technologies it will raise living standards while also bringing harms and shocks; and economists, policymakers, and technology leaders should build the capabilities to respond quickly and effectively. It is a concise version of the permissionless-innovation case inside the AI policy debate.Read more: The Geek WayOwn Your WeightsAuthor: Jamin Ball Published: July 10, 2026Jamin Ball argues that the enterprise AI debate about whether companies should “own their weights” or rent models from frontier labs is asking too narrow a question. A model weight file gives a company control over a point-in-time artifact, but not durable control over the capability stack. In his framing, the weight file is a melting ice cube: it does not get worse in absolute terms, but it falls behind as frontier systems improve and enterprise needs change.The killer detail is what Ball says companies really need to own: the data flywheel, reinforcement learning infrastructure, and evaluation harness that produce and improve the model. Simply deploying an open-weights model and declaring sovereignty leaves the enterprise with yesterday's capability and no way to compound workflow-specific learning.The pull is that enterprise AI control may be less about model ownership than operating ownership. The defensible layer is the system that turns company data, edge cases, business definitions, and evaluations into continuously improving performance.Read more: Clouded JudgementWays to Think About Token PricingAuthor: Benedict Evans Published: July 9, 2026Benedict Evans argues that today's AI token prices are a temporary signal from a supply-constrained market, not a reliable guide to long-term value capture. The open question is whether foundation models keep durable pricing power or become commodity infrastructure as data-center capacity, inference efficiency, and model competition all shift. His current read is that the visible market dynamics point toward commoditization unless something materially changes.The killer detail is the mobile data analogy. Evans says cellular networks became a trillion-dollar industry with hundreds of billions in capex after data usage exploded, but carrier stocks went nowhere because value moved up the stack. Tokens may behave similarly: an opaque unit tied to marginal cost, sold through bundles, essential to everything, yet not necessarily where profits accrue.The pull is uncertainty, not prediction. Evans lists paths to model dominance, including network effects, less competition, regulation, export controls, or a lab pulling ahead on execution, but says each requires a new fact not yet visible. Without that change, the model layer looks more like infrastructure beneath the products that capture value.Read more: SourceAlex Karp Is Saying What Every Angry CEO Is Thinking About AIAuthor: Tim Higgins Published: July 11, 2026Tim Higgins reports that Palantir CEO Alex Karp has turned corporate frustration with AI labs into a public argument about enterprise control. Palantir released a white paper, “Institutional Sovereignty in the Age of AI,” laying out steps companies and governments can take to protect themselves from OpenAI, Anthropic, and other foundation-model providers. The article links that paper to Karp's CNBC appearance, where he said “something has gone completely wrong” in the relationship between AI labs and customers and argued that enterprises are paying for tokens that create little value.The killer detail is the value-capture question. Higgins writes that Karp's critique has resonated because AI labs may gain power and insight from customer data, workflows, and decision-making, even when enterprise policies say customer data are not used for training. David Sacks amplified the concern by arguing that Anthropic is moving from the model layer into vertical applications such as science, security, legal, and coding, raising the fear that model providers will watch where value is being created and then move into those markets directly.The pull is that Karp is not alone, even if his style is unusually combative. Higgins notes that Satya Nadella has also warned that companies need to retain the learnings created when they use AI models, while Mark Zuckerberg has framed Meta's new model release partly around lower-cost frontier intelligence. The article presents Karp's campaign as one sign that established technology companies and large enterprises are trying to define where they fit when AI labs become central infrastructure, application competitors, and potential IPO giants at the same time.Read more: The Wall Street JournalThe AI Agents Are Coming for Microsoft OfficeAlex Wilhelm | Cautious Optimism | July 11, 2026Alex Wilhelm argues that one of the week's quieter AI questions is whether the productivity market that Microsoft successfully moved into subscription software is now being attacked by agentic tools. The piece begins with the infrastructure backdrop: SK Hynix raised $26.5 billion in a U.S. listing while building U.S. HBM and advanced-packaging capacity, and memory, chip, and foundry companies are now priced for sustained AI demand.Wilhelm then says the AI conversation has shifted quickly from raw capability to cost per task. He cites new model releases and vendor language emphasizing cheaper agentic and coding models, faster performance, and lower dollars per task. That matters because lower costs make it more plausible for AI systems to take on routine knowledge work at scale rather than remain a premium coding assistant market.The core of the article is Microsoft Office. Wilhelm notes that Microsoft turned Office from a one-time purchase into Microsoft 365, a large recurring revenue business with tens of millions of subscribers and a major productivity segment. Now, he says, late-stage unicorns and AI labs are pushing into the same territory: Anthropic's Cowork was reportedly used mostly outside software development, OpenAI merged ChatGPT and Codex into a tool for creating sheets, slides, docs, web apps, and long-running work, and other companies are building agentic coworkers that connect business data to documents, workflows, schedules, alerts, and apps.The article's caveat is that Microsoft has survived major platform shifts before. The argument is not that Office disappears quickly, but that the definition of office software is broadening from documents and spreadsheets into AI systems that can create, monitor, and act across workplace data.Read moreWhat Is Loop Engineering, and Who Owns It?Author: Nilesh Barla Published: July 11, 2026Nilesh Barla argues that “loop engineering” is becoming a distinct discipline because production AI agents now fail less at single prompts than at runtime: when to stop, what state to preserve, and how to recover after a bad step. Prompt engineering shapes one model call, and context engineering shapes what the model sees, but loop engineering shapes what a sequence of calls actually does.The killer detail is the three-primitives frame. Barla says a real agent loop needs halt conditions, state carryover, and recovery paths, then maps teams across five maturity levels. At the lowest level, an agent is just a model call in a for-loop with a step cap and raw history; by the higher levels, the system has structured state, explicit planning, replay, evaluation, and self-repair.The pull is organizational. If agents are becoming production systems rather than demos, someone has to own the runtime itself. The loop engineer is the role Barla gives to the person responsible for making long-running agent work dependable.Read more: Adaline LabsThe Fight Against AI Data Centers Is Just BeginningEmma Roth | The Verge | July 12, 2026Emma Roth argues that community resistance to data centers has moved from an early warning sign into a national political fight as AI facilities grow larger, more power-hungry, and more visible to nearby residents. The article starts with Apple's failed 2015 plan for a $1 billion data center in Athenry, Ireland, where a small group of residents challenged the project over noise, light pollution, flooding, traffic, and wildlife effects until Apple abandoned it in 2018.The current data-center buildout is presented as much larger and more contentious. Roth writes that residents now cite rising energy costs, water quality, noise, light pollution, and greenhouse gas emissions, while the U.S. Energy Information Administration expects commercial energy demand to surpass residential demand this year because of AI data centers and Goldman Sachs expects data-center power demand to double by 2027.The central evidence comes from Data Center Watch, which says protesters blocked or delayed at least 75 U.S. projects worth $130 billion from January to March, with active opposition groups more than doubling from 396 at the end of 2025 to 833 by the end of the first quarter of 2026. Roth also cites QTS abandoning a $12 billion Wisconsin campus, Delaware City regulators blocking a 580-acre project under the Coastal Zone Act, opposition stopping a QTS project in Prince William County, and pressure that pushed Kevin O'Leary to downsize the proposed 40,000-acre Project Stratos in Utah.The policy section describes a split between federal acceleration and local resistance. President Trump has treated data centers as part of the AI race with China and fast-tracked construction, while some Republican candidates are distancing themselves from that position ahead of midterms. Sanders and Ocasio-Cortez have proposed a moratorium until price and environmental protections exist, bipartisan lawmakers are backing ratepayer-protection measures, and states including Florida, Idaho, and Washington have passed rules on cost shifting, water use, and tax breaks. Roth's caveat is that the policy patchwork is still incomplete, leaving many communities to fight project by project.Read more6 months to live for open modelsAuthor: Nathan Lambert Published: July 12, 2026Nathan Lambert argues that open-weight AI models are facing their most serious policy test so far because U.S. officials are beginning to discuss concrete controls rather than abstract safety concerns. He says reported White House conversations about a new executive order may initially target Chinese-origin models and government use, but could create a broader review habit for frontier open models. His forecast is that a model above the capability range of GPT-5.5, Claude Opus 4.8, or GLM-5.2 could trigger a ban or indefinite delay within six months.The post separates two policy fights that are becoming intertwined: distillation and frontier capability. Lambert says the distillation campaign against Chinese models has become a form of regulatory capture because Anthropic and other closed-model companies would gain economically if Chinese open models were banned. He does not dismiss IP protection, but argues that if a closed model's capabilities are dangerous enough to justify restricting open models, the lab also has to explain why those capabilities are exposed through a queryable API. He cites unauthorized access to Anthropic's Mythos private beta as evidence that APIs are not automatically secure.The broader claim is that a unilateral U.S. ban would hurt positive actors more than bad actors if comparable open models remain available elsewhere. Lambert says the only durable ceiling would require global agreement, which does not exist, and that open models can improve safety by allowing broad inspection, adaptation, and understanding. His proposed near-term off-ramps are a strong U.S. open model release from companies such as Microsoft, Meta, or Reflection, and a broader coalition of open-source beneficiaries lobbying for safe rollout rather than prohibition.Read more: SourceAmericans Deserve a Dividend From AI Companies' RichesAuthor: Scott Stanford Published: July 14, 2026Scott Stanford argues that proposals to give the government a stake in AI companies miss the point unless ordinary citizens directly receive and control the upside. Sam Altman has discussed giving up equity in OpenAI, Washington already owns a stake in Intel, Nvidia is sharing China chip revenue, and Bernie Sanders wants large AI labs to contribute half their stock to a sovereign wealth fund. Stanford says those ideas all park value with the state, not with people.The killer detail is New Carlisle, Indiana, where AWS's Project Rainier is turning cornfields into one of the world's largest AI superclusters. The project is planned to run up to a million chips, draw more than two gigawatts of power, and represents an investment that has grown from $11 billion to $13.8 billion. Stanford uses that local transformation to argue that AI's public bargain should be visible at the household level.The pull is design. A citizen AI dividend would have to specify who earns a stake, how they hold it, and when they see cash. Without that mechanism, the AI wealth debate remains a fight over government balance sheets rather than public ownership.Read more: SourceWho Gets to Define the Frontier?Author: Mark Daley Published: July 14, 2026Mark Daley argues that Demis Hassabis is right to call for a serious institution to verify frontier AI systems, but that the power to test models is also the power to govern them. Hassabis's proposed Frontier AI Standards Body would get privileged pre-release access to advanced models, testing compute, held-out evaluations, support from national labs and security agencies, third-party auditors, and eventually authority to block models from the American market or coordinate a slowdown.The killer detail is Daley's constitutional objection. He says the proposal sometimes looks like a scientific lab, a standards body, an industry regulator, a licensing authority, and an emergency security council at once. Combining those roles because each requires technical expertise would be like putting the central bank, auditor-general, and Supreme Court in one building and calling it efficient.The pull is standard-setting. Daley's concern is not that verification is unnecessary, but that whoever writes the tests, decides what passes, adjudicates disputes, and grants market access may end up defining the frontier itself.Read more: SourceGPT-Red: Unlocking Self-Improvement for RobustnessOpenAI | OpenAI | July 15, 2026OpenAI describes GPT-Red as an internal automated red-teaming model trained to find prompt-injection vulnerabilities at a scale human red teams cannot match. The post says AI systems increasingly encounter third-party data through browsers, connected apps, local files, and tools, creating opportunities for malicious instructions hidden in emails, webpages, tool responses, or code repositories. Human red-teaming remains part of OpenAI's safety process, but the company says it is time-intensive and cannot generate enough diverse adversarial examples for model training.The system is trained through self-play reinforcement learning, with GPT-Red rewarded for eliciting valid failures and defender models rewarded for resisting attacks while still completing their tasks. OpenAI says the training environments specify threat models across settings such as local files, webpage banners, email bodies, and tool outputs. The model is kept separate from deployed production models because it is intentionally trained with malicious capabilities.OpenAI reports that GPT-Red generalized beyond its training set, including an internal replication of the indirect prompt-injection arena from Dziemian et al. (2025), where it found successful attacks in 84% of scenarios compared with 13% for human red-teamers. The post also says GPT-Red transferred attacks from simulation to a live autonomous vending-machine agent, causing price changes and order cancellations, and outperformed a prompted GPT-5.5 baseline against a Codex CLI agent on held-out data-exfiltration tasks.The article's main robustness claim is that OpenAI has used GPT-Red and predecessor models in training since GPT-5.3, with later GPT releases becoming more resistant to prompt injections. It says GPT-5.6 Sol has six times fewer failures on OpenAI's hardest direct prompt-injection benchmark than the best production model from four months earlier, that a “Fake Chain-of-Thought” attack class fell from more than 95% success against GPT-5.1 to below 10% against GPT-5.6 Sol, and that GPT-5.6 Sol fails on only 0.05% of GPT-Red's direct prompt injections. OpenAI says general capabilities and targeted over-refusal evaluations were not harmed, and says a preprint with more details will follow.Read moreAnthropic, Blackstone bet the next trillion-dollar AI business is implementation, not just modelsRebecca Bellan | TechCrunch | July 15, 2026Rebecca Bellan reports that Ode with Anthropic is the $1.5 billion AI implementation company launched by Anthropic with Blackstone, Hellman & Friedman, Goldman Sachs, and other backers. The article says the venture reflects a growing belief among frontier AI labs that enterprise adoption requires more than better models: customers need engineers who can embed inside businesses and turn AI into working systems.Ode was originally conceived by Blackstone after it used both large consulting firms and smaller AI services boutiques across its portfolio companies. TechCrunch reports that Fractional AI, an AI engineering services startup, stood out and was acquired by the joint venture shortly after the venture was announced. Fractional now forms the foundation of Ode, which has 100 engineers and works closely with Anthropic's applied AI team to identify where the technology can affect specific businesses.Ode CEO Chris Taylor tells TechCrunch that the company could someday become a trillion-dollar business if it scales without losing quality. He says an ideal customer is one whose CEO treats the AI project as a top one or two priority, whether it is a major product feature or the reworking of a core business process. Ode will operate under a “Claude-first” principle, using Anthropic technology whenever possible, but the article says it can use rival AI products when needed.The article's central implementation argument comes from Ode chief technologist Eddie Siegel, who says model selection matters but is not where most of the engineering effort goes. He compares it to the choice of programming language in software: one ingredient in a system that still has to be engineered. Bellan writes that Ode's challenge is hiring and training enough elite generalist engineers, many of them former founders, while competing with OpenAI's The Deployment Company and consulting giants that have built their own forward-deployed engineering teams.Read moreVint Cerf is working on a plan to unleash AI agents on the open internetTim Fernholz | TechCrunch | July 15, 2026Tim Fernholz reports that Vint Cerf, after leaving Google, is advising Innovation Labs on an open architecture for identifying AI agents online. Innovation Labs is a subsidiary of Identity Digital, a DNS registry company, and its proposal is to use domain-name infrastructure as part of a system for agent identity, accountability, and auditability. The premise is that agents will need a way to identify themselves if they move beyond proprietary systems and begin interacting across the open internet.The concrete proposal is DNSid, a registry that links an AI agent to an existing internet domain and uses cryptographic proofs to log its registration over time. Innovation Labs says it is trialing the standard with unnamed hyperscalers and identity companies. Cerf frames the problem around authority and accountability: what authority an agent has, where that authority came from, who is accountable for the agent's behavior, how its identity is established, and why anyone should trust it.The article's caveat is that standards are still emerging and agents are more active than static domains. Cerf says the period may be both fascinating and exasperating because the functionality is powerful and interoperability is unresolved. He compares the adoption problem to TCP/IP: competing systems may not work together until users push for functional interoperation. He also says an agentic economy is not inevitable, but that people will try to build it because delegating work to agents will be easier.Read more: TechCrunchxai-org/grok-build, now open sourceAuthor: Simon Willison Published: July 15, 2026Simon Willison argues that xAI's decision to open-source Grok Build is best understood as a trust repair move after a severe privacy failure. The CLI had triggered backlash when users realized that running it in a directory could upload the entire directory to xAI's Google Cloud buckets, including one user's reported SSH keys, password manager database, documents, photos, and videos. xAI disabled the feature, said previously retained coding data would be deleted, and released the code under Apache 2.0.The killer detail is what the codebase reveals. Willison counts 844,530 lines of Rust, only about 3% of which appears vendored, and finds remnants of the upload system still present but disabled: gcs.rs contains Google Cloud upload code, while upload_session_state() now returns a hard-coded session_state_upload_unavailable error. He also notes copied or ported tool implementations from Codex and OpenCode, prompt files, and a terminal Mermaid renderer.The pull is that terminal coding agents are becoming large, intricate software systems in their own right. The privacy failure mattered because these tools operate inside the directories where developers keep their most sensitive work; the open-source release matters because trust now depends on inspecting what an agent can see, send, and do.Read more: SourceThe Pulse: What can we learn from Bun's rapid Rust rewrite with AI?Author: Gergely Orosz and Ivan Klaric Published: July 16, 2026Gergely Orosz and Ivan Klaric argue that Bun's AI-assisted rewrite from Zig to Rust is a practical sign of how software engineering changes when models can take on large, bounded migrations with clear feedback loops. The piece does not treat the rewrite as magic: Jarred Sumner first spent hours turning design judgment into a detailed porting guide, then used adversarial review, parallel agents, compiler errors, and tests to force the work toward correctness.The killer detail is the scale. Bun had 535,496 lines of Zig, 1,448 files, and 22 million monthly downloads, making a conventional rewrite a year-long freeze the team could not justify. Using Fable, Sumner split the work across 64 agents, produced about 6,500 commits, and got the migration done in 11 days at an estimated API cost of $165,000.The pull is economic, not theatrical. If a one- or two-year migration can become an 11-day project, AI coding is not just faster autocomplete; it changes which technical debts are worth paying down.Read more: SourceOrphan risks at the frontier of artificial intelligenceAuthor: Andrew Maynard Published: July 16, 2026Andrew Maynard argues that frontier AI safety frameworks are creating “orphan risks”: harms that companies can see, but do not formally own because they are hard to quantify, do not fit catastrophic-risk thresholds, or fall outside audit-friendly compliance machinery. His target is not existing frontier safety work, but the narrowing effect that happens when private companies decide which risks count as governable.The killer detail is Maynard's contrast between measurable model dangers and threats to value. He points to Meta's three-day Galactica collapse, OpenAI's 2023 board crisis, safety-team departures, and wellbeing litigation as examples of risks that damaged trust, culture, legitimacy, or users without fitting cleanly into conventional model-risk categories. The proposed fix is an orphan-risk register: a public record of risks a company considered and chose not to manage, with reasons.The pull is accountability. Frontier developers' internal scoping choices have become a de facto layer of public governance, so the question is no longer only which risks they manage, but which risks they quietly leave outside the frame.Read more: SourceThe Lab of the Future Should Feel Like a Data CenterLatent.Space with Andy Beam and Rafa Gomez-Bombarelli | Latent.Space | July 16, 2026Latent.Space interviews Lila Sciences CTO Andy Beam and chief science officer for physical sciences Rafa Gomez-Bombarelli about the company's attempt to build an AI-run science factory. The post describes Lila's thesis as treating the lab itself as an “infinite token generator”: if internet data drove the first era of AI scaling, experimentally verified scientific data may be the next scarce training source. Lila is trying to produce that data with robotics, lab instruments, orchestration software, and AI models wired into the wet lab.The central analogy is the lab as data center. Instruments are nodes on a graph, a magnetically levitating transport layer moves materials between them, and experiment scheduling looks like a compute queue. Beam says Lila is not simply an automation company, because the point is not just throughput; it is flexibility, generalization, and experiment capture. The post says Lila has built more than 10 trillion experimentally validated “scientific reasoning tokens,” not internet text or biological sequences.The interview ranges across biology, chemistry, drug discovery, materials science, and the limits of automation. It notes that Lila rebuilt one gas-sorption measurement to run roughly 2,500 times faster, claims its general models can transfer priors from small-molecule chemistry to metal-organic frameworks for carbon capture, and describes model-suggested platinum-group-free electrocatalysts that moved from looking boring or wrong to becoming strong performers. The caveats are physical: experiments have runtimes, biology cannot always be accelerated, chains of thought can be unreliable narrators, and reward hacking becomes more dangerous when a model controls a real lab.Read more: Latent.SpaceWhy AMI Labs' Alexandre LeBrun won't call his AI “AGI” or “superintelligence”Kate Park | TechCrunch | July 16, 2026Kate Park interviews AMI Labs CEO Alexandre LeBrun about why Yann LeCun's world-model startup avoids the language of “AGI” and “superintelligence.” LeBrun says the terms are not useful because they lack stable definitions: “We never used the word AGI. And I just noticed that nobody is using it anymore; they switched to superintelligence.” His argument is that the practical frontier is not a label, but whether AI systems can understand and predict real-world states.The article explains the world-model thesis by contrasting language prediction with physical-state prediction. A large language model predicts the next word; a world model predicts the next state, such as what happens when a glass tips over. LeBrun says LLMs remain complementary and efficient for language, but the physical world is where current AI is weak. Robotics is the clearest case: hardware has advanced quickly, but robots are still brittle outside controlled routines because they lack context and situational understanding.AMI is still pre-product, but TechCrunch reports that LeBrun was in Seoul looking for industrial partners, researchers, and global companies. He says world models cannot be built entirely inside a lab because they need access to real environments. That is why South Korea appeals to AMI: robotics, semiconductors, manufacturing, and fast adoption create the kind of hardware-heavy context that software-only AI has barely touched.Read more: TechCrunchKimi K3 Tech Blog: Open Frontier IntelligenceKimi | Kimi | July 16, 2026Kimi introduces Kimi K3 as an open 3T-class frontier model aimed at coding, knowledge work, reasoning, multimodality, and long-context agentic use. The source describes the model as a 2.8T-parameter system built on Kimi Delta Attention and Attention Residuals, with native multimodality and a 1M-token context window. It says Moonshot AI plans to release model weights by July 27.The post presents K3 through benchmark and use-case sections rather than as a general product announcement. It reports results across coding, productivity, agentic, and multimodal evaluations, including DeepSWE, Terminal-Bench 2.1, Program Bench, SWE Marathon, FrontierSWE, PostTrain Bench, OfficeQA Pro, SpreadsheetBench 2, MCP Atlas, AutomationBench, BrowseComp, GDPval-AA v2, AA-Briefcase, MMMU-Pro, MathVision, BabyVision, OmniDocBench, and PerceptionBench. The source says all reported K3 results use maximum reasoning effort with temperature and top-p set to 1.0, and that different benchmark comparisons use KimiCode, Claude Code, or Codex harnesses depending on the test.Kimi's caveats are unusually concrete. The limitations section says K3 was trained in preserved thinking-history mode, so quality may become unstable if an agent harness does not pass historical thinking content correctly or if an ongoing session switches to K3 midstream. It also says K3's emphasis on long-horizon tasks can make it excessively proactive when it encounters minor issues or ambiguous intent, and recommends imposing explicit behavioral constraints for applications that require strict boundaries. The post adds that K3 remains behind Claude Fable 5 and GPT 5.6 Sol in user experience despite being competitive overall.Read moreVenture CapitalThree Years InAuthor: Tomasz Tunguz Published: July 10, 2026Tomasz Tunguz marks Theory Ventures' third anniversary by arguing that AI's central market effect is time compression. In his telling, model release cycles, company revenue milestones, enterprise adoption, and venture categories have all accelerated. Seed, Series A, and Series B still exist as financing labels, but they no longer cleanly describe company maturity when some seed rounds are larger than IPOs and the best AI companies can mature much earlier than prior software companies.The killer detail is the shift from models to inference. Tunguz argues that inference has become the dominant AI market because workloads and buyer preferences are fragmenting: video, batch, local, agentic, and real-time tasks each create different infrastructure needs. He compares this to databases splitting into OLTP, OLAP, vector, and streaming categories, with AI pushing the same specialization into inference infrastructure.The pull is that Theory sees the AI-native venture firm as part of the same pattern. The firm says it has analyzed twice as many investment opportunities with three investors working alongside a nine-person intelligence organization, using agents and research systems to map markets, source companies, and support diligence. The piece is both a market map and a statement about how venture itself is being rebuilt by the technology it funds.Read more: LinkedInVenture Has Rarely Looked More BifurcatedAuthor: Beezer Clarkson Published: July 14, 2026Beezer Clarkson points to PitchBook's Q2 report as evidence that the U.S. venture market has split into two very different realities. AI now accounts for more than 60 percent of all U.S. venture deal value, meaning the headline market can look active and well-funded even while much of the non-AI market is dealing with a much colder liquidity and fundraising environment.The thread uses that split as the setup for Clarkson's latest Origins episode with Alec Litowitz, founder of Magnetar and QStar Capital and one of Citadel's original founding partners. Clarkson says markets like this are periods of genuine uncertainty, not merely ordinary risk, which is why Litowitz's Adaptability Quotient framework is relevant.The embedded clip makes the liquidity point concrete. Litowitz says DPI is “the resolution of uncertainty” because it converts an uncertain investment into actual cash returned to LPs. In his framing, a realized dollar is a real mark, while TVPI remains uncertain until it is realized.The killer detail is the distinction between pricing risk and resolving uncertainty. Litowitz's perspective matters because QStar is a SpaceX investor and Clarkson says the conversation happened just before one of venture's most consequential IPOs. The episode's stated questions are why venture remains a way to gain exposure to innovation, how AI is changing what is investable, why liquidity is ultimately a function of time, and why uncertainty requires a different decision framework from risk.Read more: XThe Best Angel Investors in the US: Who Backs the Most Unicorns, and Who's Active NowAuthor: Ilya Strebulaev Published: July 10, 2026Ilya Strebulaev ranks angels, angel groups, accelerators, and incubators by lifetime U.S. unicorn investments, counting checks written before a company reached unicorn status. The top of the combined list is dominated by organizations: Y Combinator leads with 113 unicorn investments, followed by Plug and Play at 52 and 500 Global at 41. Sand Hill Angels is the highest-ranked angel group at 31.The killer detail is how quickly the list changes below the biggest accelerators. Strebulaev says 271 of the 304 investors in the Top 200 are individuals, or 89%. In the top 100, individuals are 91%. That makes the market underneath the large accelerator counts look much more personal: mostly operators and individual angels writing early checks from their own networks.The pull is the ranking's own caveat. Strebulaev writes that every lifetime leaderboard has a blind spot because many of the unicorns behind those totals were founded a decade or more ago, and some angels have since moved into formal funds, slowed down, or stopped investing. His post therefore separates lifetime performance from recent cohorts, including companies founded in 2015 or later and 2020 or later. For founders or allocators making current decisions, that distinction matters: a career record and a current record are not the same measure.Read more: Ilya StrebulaevAre Prediction Markets Doomed to Fail?Author: Contrary Published: July 16, 2026Contrary argues that prediction markets' current boom depends on whether platforms can prove they are more than regulated gambling with exchange-style branding. Kalshi and Polymarket have reached mass cultural, investor, and regulatory attention, but the article says the underlying idea is old: academic markets, corporate forecasting tools, Intrade, PredictIt, and other predecessors all struggled with the same linked problems of liquidity, legality, and user appeal.The killer detail is the comparison with sportsbooks. Prediction markets present themselves as peer-to-peer, transparent, and non-house-based, but sports contracts reportedly account for more than 90 percent of Kalshi trading, and the article says the platforms keep a much thinner slice of volume than sportsbooks. A market can therefore show sports-betting-scale handle while generating far less revenue.The pull is that the product's hardest problem may be distribution of wins. If a small group of sharp traders captures most profits while casual users lose interest, prediction markets may become valuable data feeds and professional tools before they become durable consumer networks.Read more: SourceRegulationExclusive: The Next Frontier of the Deportation Wars: College CampusesAuthor: Adrian Carrasquillo Published: July 11, 2026Adrian Carrasquillo reports that college campuses are becoming a new front in the fight over immigration enforcement because automatic license plate readers can turn ordinary campus security infrastructure into searchable location data. His thesis is that Flock Safety's camera network, even without direct ICE or DHS contracts, can feed deportation enforcement through local police partnerships and data-sharing practices.The killer detail is the campaign target. The Emergency Campaign to Support Higher Education, working with Schools Drop ICE, is focusing on 75 colleges and universities publicly identified as having Flock contracts. Flock says it has no ICE or DHS contracts, but activists argue the risk comes through local agencies that coordinate with federal authorities and run searches on their behalf.The pull is broader than immigration. Carrasquillo notes that license plate readers have already been abused by officers for stalking, and that Flock's AI search features can identify more than plates, including bumper stickers. A campus safety tool can become a political surveillance system when the data layer is searchable.Read more: The BulwarkThe Supreme Court Broke Independent Agencies. Here's a Way to Slow the Damage.Author: Todd Phillips Published: July 12, 2026Todd Phillips argues that the Supreme Court's decision in Trump v. Slaughter damaged independent agencies by ending for-cause removal protections, but did not leave Congress powerless. The ruling weakens the old model in which commissioners at bodies such as the FTC, NLRB, CPSC, SEC, and CFTC could be insulated from dismissal over policy disagreements. Phillips says the next fight is whether presidents can turn nominally bipartisan commissions into one-party instruments.The killer detail is the procedural fix: quorum rules. Phillips proposes that Congress require bipartisan slates of commissioners to be seated before independent agencies can act. A president

Cameron Herold sold everything he owned and has visited 54 countries in 39 months. The biggest lesson he brought back? None of it really matters. We're all just walking each other home. Cameron is the operator who scaled 1-800-GOT-JUNK to $100 million, founded the COO Alliance, and wrote the book on Vivid Vision. In this episode, he sits down with me to talk about what actually moves the needle in business and in life. In this Get Yourself Optimized episode, Cameron breaks down: ✅ Vivid Vision: how a four- or five-page description of your future gets your whole team, and your whole life, aligned ✅ The flywheel: how obsessing over one or two things built a company people called a cult ✅ R&D, or rip-off and duplicate, and why you never have to be the smartest person in the room ✅ Why the only people who should fear AI are those not using it, and how to get your team playing with it weekly ✅ The stop-doing list, and why killing off half-finished projects can free up more energy than finishing them He also gets personal about refocusing on his marriage, raising two grounded sons, and becoming the chief energizing officer in every role of his life. Tune in! The show notes, including the transcript and checklist to this episode, are at getyourselfoptimized.com/567

Who is Gwen?Gwen Acton, PhD, has built her career helping “rocket scientists”—the brilliant scientists, engineers, and technical experts behind today's most innovative companies—become extraordinary leaders. With a background spanning everything from biology to technology, Gwen was inspired by the crucial roles these experts play in industries like biotech, aerospace, AI, big pharma, and government research labs. She saw that while technical leaders are essential for organizational success, many find themselves suddenly tasked with leading teams, often without the tools or training to do so. Gwen made it her mission to bridge this gap, coaching and advising growth-stage companies and R&D groups so their top technical minds can excel both as experts and as leaders.Key TakeawaysSummaryStuart Webb and Gwen Acton PhD discuss* Why technical expertise doesn't equate to leadership proficiency* Common pitfalls technical leaders face (like micromanagement and letting go of control)* The mindset shift required to move from individual contributor to successful team leader* The most essential skills and behaviors for leadership in science and technology-driven organizations* Practical advice for influencing up, engaging teams, and leading transformation* Insights from Gwen's career and highlights from her book, “Leadership for Scientists and Engineers”* Audience questions on changing leaders' minds and the hallmarks of great leadershipWhether you're a new leader, a technical expert on the rise, or simply curious about the bridge between science and leadership, this episode delivers actionable tips and inspirational stories.Timestamps & Overview* 00:00 - Tech glitches and introductions:Stuart Webb kicks off, introducing Gwen Acton PhD, her background, and the episode setup.* 01:24 - Defining the technical leader's challenge:Gwen explains who she helps and the need for leadership skills in technical organizations.* 04:10 - Why technical backgrounds don't prepare you for leadership:Gwen highlights the crucial differences between technical problem-solving and managing people, and why logic alone isn't enough for people problems.* 07:04 - The confusion and blind spots of technical leaders:Discussion about why what worked as a technical expert often fails in leadership, and the importance of learning new interpersonal skills.* 09:16 - Letting go and the pains of delegation:Gwen addresses micromanagement, the struggle to trust others, and methods for successful delegation in high-precision environments.* 11:29 - One piece of leadership advice:Why great leaders ask more (and better) questions instead of always trying to provide answers.* 13:24 - Audience Q&A – Changing minds and influencing up:Gwen shares tips for helping leaders change direction, including the power of genuine questions and one-on-one influence.* 17:05 - Gwen's journey from scientist to leadership expert:Personal story about her path, why group dynamics and leadership in science are so important, and practical solutions for the technical world.* 20:20 - About the book “Leadership for Scientists and Engineers”:What readers will find inside and why it's designed as a practical guide for technical leaders.* 21:34 - What makes a good leader or role model?:Key traits and behaviors ranging from strategic alignment to offering feedback and motivating teams.* 26:48 - The killer question: What separates great leaders from those who plateau?:Gwen's reflections on lifelong learning, intentional skill development, and the humility to accept feedback.* 29:28 - Closing remarks and staying in touch:How to subscribe for future episodes, more about Gwen, and reaching out with questions.Resources and Links*

Today, I'm joined by Stephen Ellsworth, founder of poppi.   Originally launched at farmers markets as "Mother Beverage," poppi rebranded in 2020 and scaled rapidly before being acquired by PepsiCo for nearly $2B.   In this episode, we discuss cultivating consumer obsession.   We also cover:   How 2.5 years at farmers markets became the ultimate R&D Why the $50M revenue milestone matters Landing a deal on Shark Tank Subscribe to the podcast → insider.fitt.co/podcast  Subscribe to our newsletter → insider.fitt.co/subscribe  Follow us on LinkedIn → linkedin.com/company/fittinsider    Website: www.drinkpoppi.com/  Stephen's Instagram: https://www.instagram.com/stephenellsworth_/    -   The Fitt Insider Podcast is brought to you by EGYM. Visit EGYM.com  to learn more about its smart fitness ecosystem for fitness and health facilities. Fitt Talent: https://talent.fitt.co/  Consulting: https://consulting.fitt.co/  Investments: https://capital.fitt.co/    Chapters: (00:00) Introduction (01:49) Background and poppi origin story (02:40) Mother Beverage to poppi journey (04:50) Two-year R&D at farmers market (06:20) Shark Tank and capital inflection (08:12) Rebranding decision and strategy (11:00) 2020 timing and category alignment (12:40) Intuition-driven product development (13:40) COVID acceleration (14:45) $50M inflection point (16:20) Founder preferences: zero to one vs. scaling (17:50) Product obsession and team building (18:40) 15–20% better formula (20:00) poppi playbook and thinking differently (21:50) Democratizing better-for-you foods (25:30) PepsiCo acquisition concerns (28:45) Large incumbents vs. innovation (31:00) Post-acquisition emotions (35:15) Preferment vs. retirement (37:30) What's next (39:15) Conclusion

A24 announced a $75 million research partnership with Google DeepMind and framed it as R&D for production tools, not a content deal. Nobody cared about the distinction. Chris, Daniel, and James Blevins spend the first half of this Rough Cut on why that announcement landed like a betrayal anyway, then pull the same thread through the Paramount Skydance and Warner Bros. Discovery merger, Mark Ruffalo's New York Times op-ed on quiet industry retaliation, and what it means that four studios now control what used to take six. From there the panel turns to Christopher Nolan's press tour for The Odyssey and the practical-versus-VFX marketing dance that VFX artists roll their eyes at every single time it resurfaces. They close with a Monstrous Moonshine state of the union, walking through the June July shoot and a quick, lighthearted aside on why the show still uses AI tools like Suno for its own theme song. Mentioned this episode: A24, Google DeepMind Paramount Skydance, Warner Bros. Discovery, Mark Ruffalo, Matt Stoller Christopher Nolan, The Odyssey, D-Neg, Weta Workshop Monstrous Moonshine June July Suno Special thanks to our sponsor: Center Grid Virtual Studio: https://cgvirtualstudio.com/

For many business owners, one of the biggest missed wealth opportunities may already be sitting inside the business. Technology upgrades, software implementation, equipment, and operational improvements. These are the kinds of investments owners are already making to grow, modernize, and stay competitive. But in many cases, they may not just be expenses or deductions. They may create R&D tax credits. A deduction lowers taxable income. A credit can reduce taxes dollar for dollar. For an owner with meaningful income, that is not a small accounting detail. It is capital that can stay inside the system, be redeployed, invested, protected, or used to build long-term personal wealth. Wealth does not only leak through bad investments. It also leaks through taxes that were never strategically addressed, retirement income plans that rely too heavily on average returns, and capital that leaves the system before it ever has the chance to compound. The question is not simply how much money you make, how much you save, or whether the market performs over time. It is how much capital you actually keep, how intelligently that capital is structured, and whether your wealth plan can hold up when taxes, volatility, timing, and life events collide. Derick Van Ness is the founder of Big Life Financial, where he helps business owners turn business income into personal freedom, long-term security, and legacy through advanced tax and financial strategies. In this episode, he breaks down why R&D tax credits are not just for laboratories, tech companies, or large corporations. We also talk about the retirement problem many investors underestimate: not whether markets go up over time, but what happens when you need income during a down year and are forced to sell impaired assets. Derick walks through how volatility buffers can help protect income planning when timing, markets, and withdrawals collide.   About the Guest Derick Van Ness is the founder of Big Life Financial, where he helps business owners turn business income into personal freedom, long-term security, and legacy through advanced tax and financial strategies. Derick specializes in working with small business owners, including dentists, chiropractors, auto shop owners, and other service-based entrepreneurs who are earning well but may not yet have the systems in place to turn income into lasting wealth. His work focuses on helping owners reduce unnecessary tax leakage, uncover overlooked opportunities, and build strategies that allow more capital to stay productive. Over the course of his career, Derick has worked with more than 2,500 businesses across the U.S., identifying strategies many owners never see and helping them create stronger financial outcomes. His perspective is especially useful for business owners who want to move beyond simply making money in the business and start using the business as a vehicle for personal freedom, wealth preservation, and legacy. Get an R&D tax credit estimate: https://biglifefinancial.com/money-school-credits Learn more about Big Life Financial: https://biglifefinancial.com/       About Your Host From pro-snowboarder to money mogul, Chris Naugle has dedicated his life to being America's #1 Money Mentor. With a core belief that success is built not by the resources you have, but by how resourceful you can be. Chris has built and owned 19 companies, with his businesses being featured in Forbes, ABC, House Hunters, and his very own HGTV pilot in 2018. He is the founder of The Money School™ and Money Mentor for The Money Multiplier. His success also includes managing tens of millions of dollars in assets in the financial services and advisory industry and in real estate transactions. As an innovator and visionary in wealth-building and real estate, he empowers entrepreneurs, business owners, and real estate investors with the knowledge of how money works. Chris is also a nationally recognized speaker, author, and podcast host. He has spoken to and taught over ten thousand Americans, delivering the financial knowledge that fuels lasting freedom.   Resources Private Money Guide:  https://go.moneyschoolrei.com/book-podcast Wealth Wednesday Webinar: https://go.moneyschoolrei.com/wednesday-webinar-podcast Mapping out the Millionaire Mystery:  https://go.moneyschoolrei.com/newbook-podcast    

Dr. Z and Dr. C sit down with Dr. Sarah Gheuens, Chief Medical Officer and Head of R&D at Agios Pharmaceuticals, to break down one of the most frequently heard but least understood terms in the treatment conversation: FDA approval. From the phases of clinical development and the role of regulators, to the difference between traditional and accelerated approval pathways, Dr. Gheuens demystifies the science and the process in plain language. They also discuss why asking questions of physicians about their care is one of the most important things a patient can do. SHOW DESCRIPTION Cheat Codes is intended for patients, caregivers, providers, and the greater community of people who are impacted by Sickle Cell Disease.  Each episode, Cheat Codes strives to provide listeners with critical education, the latest scientific updates, and voices from the Sickle Cell community.   Join an inclusive community and build connections with other hemolytic anemia allies by following @AllyVoicesRising on Instagram. TRANSPARENCY STATEMENT  Cheat Codes: A Sickle Cell Podcast is made possible by Agios Pharmaceuticals Inc. Visit Agios.com to learn more. The following Agios-supported programs are intended for informational and educational purposes only and are not intended as medical advice. Please speak with your healthcare professional before making any treatment decisions. Hosts Drs. Ahmar Zaidi and Mike Callaghan are Agios employees.  

In this episode of The Capital Raiser Show, Richard C. Wilson sits down with Marc Halpern, co-founder of the Deep Due Diligence Investors Club, for a fireside chat on screening private placements at scale, managing risk in alternative investments, team-based due diligence, and what separates disciplined investors from everyone else. Marc shares lessons from five decades of breakthrough R&D and private investing, including the "jockey, horse, and track" framework, how his club uses AI to screen 400+ deals a year in seconds, and why reading every word of a PPM is still non-negotiable. The conversation dives into deal screening discipline, portfolio strategy, diversification, stress testing sponsors through market cycles, and how a team of 58 investors from wildly different professional backgrounds covers each other's blind spots better than any single expert could. Topics covered include: The "jockey, horse, and track" framework for evaluating any deal How AI deal screening tools cut through hundreds of opportunities in seconds Deep due diligence in teams and why diverse professional backgrounds matter Minimizing risk vs. eliminating it - and why you can only do one Portfolio strategy before individual investment selection Diversification vs. de-worsification and finding the right balance Stress testing sponsors through the Fed rate cycle of 2022-2023 When to walk away quickly and take no for an answer The biggest misconception private investors have about succeeding in alternatives The Capital Raiser Show brings together billionaire investors, family offices, elite entrepreneurs, and capital allocators to discuss investing, scaling, strategic growth, and wealth creation. Subscribe for more interviews with top investors, founders, family offices, and industry leaders.

Sam, Dylan, and Dark Smith are back to break down: Dark Smith's job interview to become an escort and the loofah-color swinger code of Florida's villages, the Boyle Heights fire spilling Freon and ammonia while firefighters spray a blaze they can't reach (and the owner being a Karen Bass donor), a documentary trying to abolish the Electoral College, the deep origins of the Vatican built on top of Vatican Hill, the pagan goddess Cybele and the Anatolian mother god Magna Mater brought to Rome via the prophetic Sibylline Books after Hannibal's slaughter at Cannae, the blood-baptism bull-sacrifice rituals later dug up by archaeologists, H.P. Lovecraft's "The Rats in the Walls" and its cannibal cult, the serpent-shaped Vatican auditorium and the Pope's fish-head mitre tied to the Babylonian fish god Dagon (who turns out to be Godzilla), Sam's pitch for roller derby as the next great pro sport, the Iran war being "over" again with IAEA inspectors as the "milestone" achieved by undoing the milestone, Israel allegedly trying to assassinate the negotiators before Trump called them off, Nicki Minaj's distraction-death theory, JD Vance's Thiel-engineered glow-up, the Albanians tearing their country apart over Jared and Ivanka's island and Sam's theory it's about controlling the Strait of Otranto, the nu-metal psyop, military stock buybacks instead of R&D, Bezos prioritizing AI water over "baseline human comfort," and a woman who lost ten years of memory from straining too hard on the toilet. Subscribe and give us that sweet brown hype.   Grab Tickets To Sam Tripoli's Live Shows At: https://samtripoli.com/events/   Miami, Fl: 7/31-8/1 Lawerence, KS: 9/17-9/19 Tulsa, OK: 10/9-10/10 Dallas, TX: 11/07 New Orleans, LA: 11/13 - 15 Austin, TX: DEC 11th-13th:   Buy Our Merch or Sam Will Fight You: https://conspiracy-social-club-aka-deep-waters.myshopify.com/   Subscribe to the Patreon: https://www.patreon.com/AkaDeepWaters   Check out Dylan's instagram - @dylanpetewrenn   Check out Deep Waters Instagram: @akadeepwaters   Check out Bad Tv podcast: https://bit.ly/3RYuTG0   THANK YOU TO OUR SPONSORS:   LUCY.CO/CSC Promo Code "CSC" to get 20% off your first order   HIMS.COM/CSC HIMS.COM/CSC for your FREE Online Visit

If your career looks successful on paper but feels off inside, you're not alone. Many mid-career professionals reach a point where the work that used to fit no longer does. Breanne Gjurich-Wozniak's story is one path forward: a portfolio career that combined her biotech expertise with her animal welfare interests. Breanne had a career that looked solid from the outside. Biotech R&D, then regulatory work submitting new technologies to the FDA. Director-level. Real success. Inside, the fit was changing. The work was drifting.She started looking at R&D and regulatory roles she was qualified for. Nothing gave her any excitement. That's when she found HTYC's free 8-day mini course and joined the Career Change Bootcamp. She built one 5-year career plan with her manager and a separate one with her career coach. The two plans were so different. The energy in each was so different. She knew it was time to leave. What followed wasn't a traditional job search. It was 20 conversations with strangers, a series of career experiments, and surprising feedback from people who'd only met her for an hour. Her coach introduced the idea of a portfolio career: multiple types of work, multiple interests, all under one umbrella. Breanne built a consulting business that combined her biotech expertise with the animal welfare work she'd always wanted to do. What you'll learn: The 5-year career plan exercise that reveals when your current path no longer fits How to use Social Goldilocks conversations to get strangers to see what you can't about yourself What a portfolio career actually looks like (and why one full-time job isn't always the right structure) How to translate decades of industry experience into consulting work Why "successful on paper" is the hardest career trap to leave Our book, Happen To Your Career: An Unconventional Approach To Career Change and Meaningful Work, is now available on audiobook! Visit  happentoyourcareer.com/audible to order it now! Visit happentoyourcareer.com/book for more information or buy the print or ebook here! Want to chat with our team about your unique situation? Schedule a conversation Free Resources What career fits you? Join our free 8 Day Mini Course to figure it out! Career Change Guide - Learn how high-performers discover their ideal career and find meaningful, well-paid work without starting over. Related Episodes Changing Careers (When You Don't Know Your Next Job Title) (Spotify / Apple Podcasts) How to Figure Out What You Really Want (Spotify / Apple Podcasts)  

Turning activewear into technology that actually supports how you move — that's the magic

This week we talk about LDL, HDL, and cardiovascular issues.We also discuss one-time therapies, statins, and pharmaceutical economics.Recommended Book: Blood by Dr. Jen GunterTranscriptCholesterol is the most common type of what's called a sterol, which is a type of steroid, but also structurally technically an alcohol. But functionally, and classified by scientists, cholesterol is a lipid, which in this case is similar to a fat in all but how the body uses it. Cholesterol is the type of sterol most commonly found in animals—other types are found in plants and fungi—and its function, and this is where it varies from fats, which are used to store energy, is to basically help hold the cell membrane together, and it also serves as an intracellular messenger.Cholesterol is especially prevalent in the brain and spinal cord of animals, but it's found throughout their bodily tissues, as well, and again, it's vital for holding everything together and helping things communicate, in addition to being a precursor for vitamin D, steroid hormones, and bile.You want to have cholesterol, then, as without it you would be dead.Too much cholesterol in the blood, however, can also make you dead, especially when it's bound to what's called low-density lipoprotein, or LDL, as that contributes to cardiovascular disease like heart attacks and aneurysms, which can massively impact one's overall wellness and quality of life, and at extremes lead to the whole system shutting down as a consequence of heart attack, stroke, and the like.A lot of things can contribute to the development of cardiovascular disease, including habits like smoking, genetic predisposition, and the enthusiastic consumption of alcohol and unhealthy foods. But high blood cholesterol, of the LDL variety, is one of the top contributors, as these low-density clusters of lipoprotein can clog the pathways that blood takes throughout our bodies. Other, denser types of lipoproteins, HDLs, can clear it, like a heavier, denser substance pushing through clogs of less-dense materials that are gumming up a pipe, but LDL is at times accumulated as a result of consuming delicious but unhealthy foods, which are hard to avoid, and for some people the only consistently available and affordable foods; and for other people LDL accumulates as a result of their genetic predispositions—two things that are devilishly difficult to change.What I'd like to talk about today is a new type of therapy that may be very good news for people who struggle with the accumulation of LDL, and why this is being seen as very good news more broadly, at the scale of entire nations, as well.—Pharmaceutical company Eli Lilly is testing a new, experimental drug called VERVE-102 which is a one-time infusion that is currently administered over the course of about four hours, and once completed, it turns off a gene called PCSK9, which is responsible for making a protein that regulates cholesterol levels in humans.As I said, this drug is still being tested, so these are early results. But in a study of 35 people with high cholesterol levels, high levels of LDL or LDL-C, which is short for lipoprotein cholesterol, they found that this infusion, which again, is a one-time treatment, so get it once and then theoretically at least you never have to get anything done ever again, it reduced those LDL and LDL-C levels by as much as 62%, and that reduction was maintained a year and a half after the infusion; that's how far out they're retested so far, and the hope is that each retest will continue to show the same.On the strength of those very promising results, a Phase 2 study has been planned by the end of 2026, and the US Food and Drug Administration, the FDA, previously fast-tracked this existing study, because of the promise and potential this drug already demonstrated in early studies; all of which is considered to be very significant progress and possibility.To understand that significance, though, it's useful to know some health stats. And I'm going to focus on the US here, as that's where this drug is being developed, but many wealthy countries have similar stats, at least in terms of cardiovascular disease struggles.As of 2024, which is the last year we had good, cohesive data on this in the US, it was estimated that about 11-12% of the US adult population has high cholesterol levels. This typically doesn't come with any symptoms, but it can contribute a higher risk for all those cardiovascular diseases, including heart attack and stroke. A further 86 million US adults have borderline or elevated cholesterol levels, which can easily tip higher, but also, even in that existing, elevated state, contribute to negative cardiovascular outcomes.There are treatments for high cholesterol, the most common of category of which are called statins, which reduce the production of LDL by inhibiting an enzyme that produces cholesterol in the body.Unfortunately, these drugs do come with some usually minor side effects, which can cause patients to stop using them, and they have to be taken daily, ideally at the same time each day. That necessity for consistency leads to a lot of incorrect or incomplete usage, which reduces the effectiveness of these drugs. But it's also estimated that only about 54.5% of US adults who would benefit from statins are currently taking one—so that's people who could benefit and who have it prescribed, and then within that number are all the people who are taking this drug incorrectly or incompletely, reducing the effectiveness. So a relatively small number of people who should probably be on these things are getting the full benefit they offer because of the nature of the drug.And that's not great, because in the US alone, heart disease is the leading cause of death for pretty much every adult demographic; men, women, people of most racial and ethnic and economic groups, you name it, heart disease is the biggest threat to their lives.One US citizen dies every 34 seconds of some kind of cardiovascular condition, and as of 2023, 1 in every 3 deaths in the US was caused by the same, adding up to just over 919,000 people that year.Between 2021 and 2022, alone, the cost of services and medications related to heart disease added up to more than $168 billion; again, that's just in that period, and just in the US.And once more, these are ailments that are caused or heavily influenced by high levels of cholesterol, which are themselves amplified by common lifestyle choices, environmental factors that are hard for many people to avoid, and just by raw, dumb luck because of genetics.This treatment category, then, is being seen as a pretty big deal because a one-time infusion means those who receive it don't have to remember to take a pill every day at the same time, and won't experience those statin-based side-effects.It also means that people who are currently costing the medical system a bunch of money each year, because they need treatments for all the issues they suffer as a result of high cholesterol, will suddenly cost the system a lot less money, for treatments and medications. Not for nothing, their health and quality of life will likely improve as well. So in addition to having better, healthier outcomes personally, their cost to healthcare systems will drop.Eli Lilly's drug isn't the only one currently working its way through clinical trials, either.Amgen is working on a similar treatment, and Novartis and Ionis Pharmaceuticals have drugs that are even further along in the process, their medicines that cut heart attacks, strokes, and cardiovascular deaths could be approved by the FDA as soon as next year.There are a lot of caveats worth noting here, including that the science is still out as to whether this approach, silencing proteins that lead to the creation of more LDL and a similar substance called Lp(a)—which is more dangerous because it's stickier and thus more likely to get stuck in important blood pathways, and it's also more likely to be caused by genetics than lifestyle—the word is still out on whether reducing these things in the body actually reduces hearth attacks and stroke.Some people have had this particular risk variable dramatically reduced, but have still suffered from cardiovascular events, which raises the question of whether this path is the right one to take in trying to reduce this category of health issues; the correlation between LDL and heart attacks and strokes might not be a clear-cut as long assumed.There's also the issue of price. Drug-makers are economically incentivized to sell treatments over cures, because that means they can continue selling their product over time, potentially for the life of the patient, and a cure, in contrast, is a one-time hit that in theory should alleviate the need for future treatment.There's a chance, then, that the drug-makers will decide they need to make these one-hit treatments really, really expensive in order to make their R&D dollars back and to make the kinds of profits their investors expect from them. That could then reduce the potential audience for these treatments, even if they are effective, and could further slow their deployment and future research in this space.If these trials continue to go well, though, there's a good chance that this combination of similar but distinct treatment types will provide a more sustainable alternative to current options, and that, like the recent bogglingly rapid and widespread deployment of GLP-1 treatments for all sorts of issues, could lead to a new paradigm in this facet of the medical world.Show Noteshttps://en.wikipedia.org/wiki/Cholesterolhttps://en.wikipedia.org/wiki/Cardiovascular_diseasehttps://en.wikipedia.org/wiki/High_cholesterolhttps://pmc.ncbi.nlm.nih.gov/articles/PMC10982736/https://www.cdc.gov/heart-disease/data-research/facts-stats/index.htmlhttps://www.who.int/health-topics/cardiovascular-diseases#tab=tab_1https://www.ama-assn.org/public-health/chronic-diseases/what-doctors-want-patients-know-about-high-cholesterolhttps://en.wikipedia.org/wiki/Statinhttps://pubmed.ncbi.nlm.nih.gov/42187087/https://abcnews.com/GMA/Wellness/new-drug-game-changer-people-high-cholesterol/story This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit letsknowthings.substack.com/subscribe

In this conversation, host Tad Hussey (KIS Organics) and Dr. Deron Caplan (CannaCribs Horticulture Consulting) discuss the intersection of agricultural science and commercial cannabis cultivation, focusing heavily on irrigation, lighting, and research methodologies. Drought Stress vs. Drybacks: Dr. Caplan clarifies a major industry misconception by distinguishing standard irrigation "drybacks" from actual "drought stress." He explains that drybacks work primarily by pulling oxygen into the root zone (which cannabis loves), whereas true drought stress requires pushing the plant past the point of water availability. The Nuance of Cannabinoid Bumps: While Dr. Caplan's landmark PhD research found that controlled drought stress in week seven of flowering could boost THC/CBD content by 30% to 40% without losing yield, he notes that repeating this in smaller pots with faster drybacks yielded no positive results. This highlights just how incredibly nuanced and difficult it is to trigger beneficial plant stress without harming the crop. Organics vs. Mineral Salts at Scale: The two debate the logistics of commercial cultivation. While mineral salts offer strict baseline consistency and easier pathogen sterilization for medical export markets, Dr. Caplan notes that data-driven, evidence-based living soil systems have come a long way and are proving to be increasingly scalable. Debunking the Leaf-Tip Cloning Myth: Dr. Caplan shares that his early research on propagation disproved a ubiquitous legacy market myth: cutting the tips off clone leaves actually reduces rooting success, unless leaves are so large that they are actively shading adjacent clones. Advanced Canopy Management: They discuss the massive yield and quality benefits (15% to 20% increases) of under-canopy lighting (UCL). Dr. Caplan details how commercial facilities can use a PAR meter to calculate the Leaf Area Index (LAI) to mathematically standardize how much foliage to prune, rather than relying on visual guesswork. The Future of Research: Dr. Caplan highlights his new commercial R&D facility in Canada designed to test applied science practices. They close by discussing the validity of peer-reviewed data versus private white papers, concluding that regardless of a grower's background, scientific data is the ultimate tool for bridging the gap between organic and conventional cultivation. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

Have you ever lost the joy in your creative work — that sense of fun you had when you were starting out, before the admin and the algorithms drained it away? How do mid-career creatives get it back, and what can a four-year-old teach us about play? Austin Kleon talks about productive procrastination, silly rituals, the case for paper reference books in an AI world, and how his newsletter went from a marketing cost to the day job that keeps the lights on. In the intro, Does social media still sell books? [Self-Publishing with ALLi]; Trial by algorithm [The Bookseller]; Publishing's AI Hypocrisy Problem [The New Publishing Standard]; ALLi AI survey for authors; Brave New Bookshelf Podcast, and Pics from signing at BookVault. Today's show is sponsored by ProWritingAid, writing and editing software that goes way beyond just grammar and typo checking. With its detailed reports on how to improve your writing and integration with writing software, ProWritingAid will help you improve your book before you send it to an editor, agent or publisher. Check it out for free or get 15% off the premium edition at www.ProWritingAid.com/joanna This show is also supported by my Patrons. Join my Community at Patreon.com/thecreativepenn Austin Kleon is the New York Times and international bestselling author of nonfiction books, including Steal Like an Artist, Show Your Work!, and Keep Going, as well as an artist, professional speaker, and poet. His latest book is Don't Call It Art: 10 Ways to Create Like a Kid Again. You can listen above or on your favorite podcast app or read the notes and links below. Here are the highlights and the full transcript is below. Show Notes Why Austin wrote Don't Call It Art now, and what his kids taught him about creative joy Productive procrastination, silly rituals, and treating writing like Lego Comedy as a philosophical position, and giving yourself permission to be bad in private Sharing process in the algorithm era, and why your whole life is the process Bibliomancy, paper reference books, and what AI can't give you that a dictionary can Style, the Taco Bell distinctiveness rule, and how Austin's newsletter became his day job You can find Austin at AustinKleon.com. Transcript of the interview with Austin Kleon Jo: Austin Kleon is the New York Times and international bestselling author of nonfiction books, including Steal Like an Artist, Show Your Work!, and Keep Going, as well as an artist, professional speaker, and poet. His latest book is Don't Call It Art: 10 Ways to Create Like a Kid Again. So welcome back to the show, Austin. Austin: Thank you for having me back. It's nice to talk to you again. Jo: You were on the show in March 2020, and at the time, your book was Keep Going, which was prescient considering the pandemic and politics. So I wondered, why this book, Don't Call It Art, now? Was this something you see in the creative community or your own life that made you want to write this book? Austin: Keep Going is a book about what happens when the world goes crazy around you and you're still trying to do your creative work. This is a book about what happens when inside has bottomed out. Keep Going is a book about the world bottoming out, and you're worried that your own creative work is going to bottom out too. How do you keep pushing through and keep making stuff? This book, to me, is about what happens when you bottom out inside—when you've lost that love and feeling for the thing that you wanted to do, and you're just not connecting with it in the way that you used to or the way that you want to. How do you get back? How do you return to that sense of joy and wonder and fun that we have when we're starting out? And for me, it was being around my little kids that taught me how to tap into that. My kids were natural—they didn't have any creative hangups. I would spend all day talking to people who had creative hangups, and then I'd get back in the house, and I'd just be around these beings who didn't have any of them. It was really instructive. I felt like, if I could bottle the energy of my kids when they were about four years old and try to put it in a book, I think it could really help a lot of the people that I run into, and the people with the kinds of problems I hear from. Jo: You mentioned bottoming out. How do people know when they've hit that point? Austin: You just don't want to do it anymore. You're kind of like, “This just isn't giving me back what it used to.” When we start with our creative work, that's the thing that juices us. We come away from it feeling full up. I think you hit a certain point where you start to feel drained after it. Or maybe you don't feel drained by the thing itself that you're doing—maybe it's all the stuff around it, which is more often the case. For example, if you're a mid-career writer like me, who's been publishing books for 16 years now, I still really like writing. I still really like drawing. I still really like cutting and pasting and putting things together. It's the admin around the work—the emails, the meetings, the running-a-business part of it—that's super draining for me, and that stuff can start to bleed over into the creative work. So it's really important for me to make sure that I'm having some playtime, some R&D, some research and development time, to make sure it's not just all business. When you take the thing that you love and you turn it into the thing that you make a living from, you can really run into a lot of problems. Jo: I'm at 20 years, so I know exactly what you're saying, and a lot of listeners are the same. We love writing books, but it's all the stuff that goes around it. So for those of us who do this for money as well as passion, what are some practical ways to have more fun with our creativity? Austin: Something I learned from my kids is that you really are your most creative when you're supposed to be doing something else. So one of the things I use a lot in the studio is productive procrastination. Whatever I'm supposed to be working on, I start another little project, and that's my little naughty fun time. When I first come into the studio, I try to do something that I'm not supposed to be doing—something that I won't have much to show for. That could be making one of my blackout poems. That could be making a collage in my notebook. It could also be sitting here. I have a bass in the studio now, so I can practise my bass guitar. Sometimes I'll do that for the first 15 minutes just to get in that headspace of, “Hey, what's it like to do something just for yourself? Just because you want to do it?” The juice that you get from that little naughty “I'm going to do what I'm not supposed to be doing right now” thing, that carries into the rest of the day. It's like a nice start to things. Jo: Do you think that play could be something different to what we make our money with? For me, writing novels and stories is great fun in one way, but it's also what I then publish and make money on. So writing stories is more serious, I guess, than playing with Lego or something. Austin: Right. So the trick is, how can you make writing your stories like playing with Lego? That's kind of been my whole career. I hate staring at Microsoft Word and that blinking cursor, taunting you like, “Come on, what have you got?” A lot of my creative life has been about trying to make it more playful, trying to make it feel more like a game. That's how I came up with my blackout poems. I take an article from The New York Times and I black it out until it only has a few words left behind. It sort of looks like if the CIA did haiku, for some people listening. That was one little exercise. Then weirdly, that side thing that I thought was just play, just fun—that turned into my first book. So then it's, okay, what else can I mess around with and play with? I do a lot of collage work in the studio, and I rarely actually use that for any of the books. Sometimes I use it for my newsletter to illustrate the newsletter. But it's always about trying to figure out, how can I make writing a game? How can I make it more playful? There are different things that I do to make it feel more playful. One of them's really stupid. I really believe in silly rituals because I think silliness is really powerful. People talk about their daily rituals—Mason Currey has that great book, Daily Rituals: How Artists Work. When I was reading that book, I realised it was really the silly stuff that I really liked. There was, I think it was Balzac counting out coffee beans or something before he got to write. Or Steinbeck sharpening 12 pencils or something goofy like that. So one of the things I like to do before I write is that I have these cigarette pencils. They're pencils that look like cigarettes in the studio. I put one in my mouth before I start writing, and I pretend to be some old '40s writer on a typewriter. I like doing goofy stuff in the studio because I think when you do goofy stuff—stuff that you'd be embarrassed if anyone else saw it—it gets you in that playful state. Jo: It's interesting. In your book, you have a section that says, “Don't take things too seriously.” For many of us, we write memoir for example, and that is very close to us. It's like the deepest expression of what we want to say in the world. It feels very serious. So how can we hold things more lightly and not take things so seriously? Austin: For me, comedy is actually a philosophical position. What I mean by that is, I think a lot of people set out with a tragic model of creative work. They think, “Oh, I have this special gift,” or, “I have this thing that I really need to do, and I need to put it out into the world, and I need to make the world look more like I want it to look.” They have this idea that, “Through blood and sweat and tears, I'm going to see this thing through, and I'm going to push it into the world, and I'm going to have my way.” I think there's another way of working where it's more like, “I'm just a normal person trying to play with my environment, and take my experiences and put them into something interesting. So I'm going to play and use my wits, and we're going to see what we come up with.” Those really are two modes of life. The pandemic taught me that it was really when we were keeping our sense of humour, when we were having a laugh and keeping our egos in check around the house and just acknowledging how goofy we all were and how ridiculous the situation was, that seemed to be when we were really thriving. Versus, “Well, we're in this tough situation. We've got to make it into what we want it to be.” That felt really bad. But when we cruised along and we were just improvisational, when we went at things with a kind of lightness, that worked. There's a great Italo Calvino essay about lightness in Six Memos for the Next Millennium. Lightness is really underrated. Even when we're going about heavy work, having a sense of lightness and play with it just makes the work better. That's a philosophical position of mine. I aspire to comedy. I aspire to a comic outlook on life. I'm just a creature with a body who's going to die, and I'm fundamentally ridiculous. Life is pretty absurd. You just make the best of it. Jo: There's certainly some truth there. Staying on a similar theme, you have a chapter in the book on permission to be bad. Many of the listeners also have your book Show Your Work, and it shaped many of us into sharing our work in progress. It feels quite dangerous now, in a world where judgment is much louder than it maybe was when you wrote Show Your Work. So tell us a bit about permission to be bad versus should we keep some of this private? Austin: Permission to be bad is about the making part of things. It's the private part. It's permission to be bad when you're in private, when you're actually doing the work. Show Your Work is a book about what you do after you've done the work, or while you're doing the work. It was never about putting up a webcam and running a 24/7 feed. It was more like, hey, what are the ways that I can connect with the kind of audience I can build while I'm making the work itself? So the way I see permission to be bad is, you really have to give yourself permission when you're not sharing, when you're off screen, to really be as bad as you want to be. It doesn't necessarily mean quality-wise. I think it also means letting yourself write stuff that you would never say on social media. Letting yourself read stuff that you wouldn't admit you were reading on social media. Letting yourself listen to stuff. Letting yourself really be that unfiltered, unhinged, private person that you want to be. Then when it comes to sharing, you put some time in between that input time, that making time, and the sharing time, and then you share what you think is going to be useful or helpful or interesting to other people. Jo: I think you wrote that book before TikTok, and how fast people are moving. Do you think people need to slow down a bit in what they share, maybe? Austin: I don't know. I obviously had a lot more faith in social media back then. I use all the principles from Show Your Work in my newsletter. Newsletters are very much the new kind of great thing. They're doing a lot of the work that social media used to do, in that you're still able to have this direct connection with the people that you're trying to reach. The big problem with social media now is that it's all algorithmically tuned, where the people that are following you don't see the stuff that you're doing most of the time. What you have to do now, if you want the people who are following you to see your stuff on social media, is you have to make stuff that the algorithm likes. That's a whole different thing. As far as the Show Your Work principle—which is share your process as much as your product—that carries over to any platform. In my newsletter every Friday, I share a list of 10 things that were going on behind the scenes here. It might have been what I was watching on TV, what I listened to, a new pen I was trying out, or something like that. The Friday newsletter is almost always process stuff. When I talk about process, my definition is actually very broad. For a lot of people, it's drafting, editing, whatever. For me, the process is the whole life. The process is almost everything except the finished thing. A writer's life is 24/7. My friends who have real jobs really are like, “What do you do all day?” And I'm like, “Well, what do you mean?” They're like, “Well, I see you out on your bike ride.” I'm like, “Yes, when you see me out on a bike ride, I'm thinking through something half the time.” If I'm watching TV, I'm thinking, “Hey, would this be good in the newsletter?” I'm never off. My whole life—everything is copy, as Nora Ephron said. That's part of the job. It's very hard to turn off. So I see the whole life as process, and the question becomes, what little bits and pieces of that life and that process can you share with people while you're making the things that you hope to sell them later? Right now, I'm in a cycle where I'm selling this book, but all these people have showed up because I've shared my process every week for the past seven years since I put out a book. Jo: It's funny you say that. I was at the dentist yesterday, and— My dentist literally asked me, “So where do you get all your ideas?” This is a common question for all of us, right? And it just becomes so hard to explain that to people who don't walk around in the world just constantly getting ideas. Austin: I can't believe I'm going to tell this story. I was getting my vasectomy after my second kid, and I was talking to this doctor just before the operation. He said, “So what do you do for a living?” I said, “I'm a writer.” He said, “Oh, that must be cool. You get to use your brain.” And I said, “That's everything that you want your doctor to say.” I was going to say, “Please use your brain,” before he's about to cut into you. He said, “Oh, no, no. What I mean is, I know what I'm going to do every day for the next 10 years.” He knew exactly what his day was going to look like. He said, “You have to use your brain. You've got to figure out new stuff.” I was like, “Oh, that's really interesting.” That's the trade-off, right? He's got the job security. He knows what he's going to do. Every writer has a moment where they have to talk to a normal person about what you do. Jo: I was going to say, I'm married to one. Austin: Now, my wife, on the other hand, grew up the daughter of a writer, so she knows exactly what it's like. Nothing ever phases her. She's totally used to it. She's used to me staring off into space, completely checking out of a conversation. She's used to me using lines on her that I'm going to put in a piece later. She's used to the whole rigmarole. It's very handy. I've been very lucky in that sense. Jo: Coming back to the book, you talk about your use of bibliomancy for inspiration. Since we're talking about that, tell us about it. I think all the book people listening will be happy. Austin: I'm a person who still keeps a dictionary nearby—a paper dictionary. I keep a big old American Heritage. It's just a big, thick book. When I really don't have any ideas, I will turn at random to the dictionary, close my eyes, stick my finger down the page, open my eyes, and just see what I come up with. Sometimes just that act will give me an idea. I also do that with books. I'll go around the studio, pick up a book, flip to a random page, and just see what it says there, or read an old piece of marginalia that I've left in a book. I believe deeply in the power of bibliomancy, and I think it's a case for paper books. I'm one of those people that still really believes in reference books. I've started collecting more and more of them. I have an old, big dictionary that's always open on my desk, and I look up words. I learned from John McPhee, the writer, that you should look up words that you think you know. That was the first time I'd ever heard anyone say that. So I look up words that I think I know. Instead of reaching for a thesaurus when I need a different word, I actually just look up the definition of the word that I already have. That's another McPhee tip. The other thing that happened that I thought was really interesting is, I got a Roget's for the first time—a thesaurus. I don't think most people know what an actual thesaurus is. Most people think of a thesaurus as a synonym finder, and that's not actually what a thesaurus is at all. A thesaurus is more like an encyclopaedia, weirdly. You look up things based on big concepts, and then it gives you a bunch of words to look up later. It's a very strange thing. It's not what most people think it is. I have a couple of editions of Roget's in here. I like the really old Roget's from the 1900s because they actually have opposing ideas facing each other on the page. Do you have an old-school Roget's? Have you ever looked through one? Jo: I don't have one now, but I certainly grew up with them. I was literally just thinking, I wonder if there are ones for Americans and ones for British people, because so often we say different things and mean different things. I always hear Americans say, “Oh, that's a doozy,” or something, and it means the complete opposite thing here. Austin: Like if you say “fanny pack” over there. That means something very different than it means here, right? Chips or fries, that kind of stuff. So I wonder if there are different ones for different cultural references. Jo: I don't know. Austin: As people, with ChatGPT and all these LLMs and stuff, people are like, “Why would you ever pick up a paper reference book?” And I'm like, “I actually like the friction.” I like having to move in space and go over to my dictionary. I like flipping the pages. I like having to scan a page for the word I'm looking for, because— This marvellous thing happens when you're looking for the word, where you bump into all these other words. If you're a word nerd, you get to start thinking about the root of the word—oh, why is this word next to this word? Well, it's because they share the same root. Then you're going down all these fun rabbit holes. The thing that I'm trying to do as a writer and a creative person is, I'm trying to get to the thing that I didn't know I was looking for. The thing that people misunderstand about AI, I think personally, is that it's a great tool if you know what you're looking for. If you're like, “Find me this thing. I want exactly this. I want to see a picture of a dog wearing a king's costume,” or some crap like that, then it can spit that picture out for you. Or, “I want to know what happened on this day,” and whatever. It can do that. But that's not actually what I'm doing most of the time when I'm writing or making something. I start with an idea, but what really happens—the magic of writing and the magic of making stuff in general—is when you discover something that you didn't even know you were headed for. That's the real magic for me. Sometimes I have an idea and I want to articulate it for people, but more often than not, there's something that bothers me or something that I want to talk about, and I sit down and write, and I figure out what it is that I actually have to say and what I actually think. Every writer really knows this, and that's why the dictionary, stuff like that, those are ways of training you to get in that discovery mode. “Well, let me—oh, I bumped into this. I went looking for this one thing and then I ran into this other thing.” That's why I love the library. I don't know what system you use over there, but you look for one book in the Dewey Decimal System over here, and then, okay, here's all these other weird books next to it. Then you end up with three other books other than the one that you were looking for. That's the magic. To me, that's the magic of creative work, discovering what you didn't know you were looking for. That was particularly important for me when I was writing this book because we discovered that my wife has a condition called aphantasia. It's very rare in the population, about 2 to 3% of people. There's probably some people listening to this right now who are like, “What is this? Tell me.” Jo: Aphantasia actually more common in the creative industries. Austin: Yes. What it is, is that you don't see—when I say close your eyes and picture an apple, you don't actually see the apple in your head. You can think about an apple and the qualities of an apple, but you don't actually see it. Some people, and it's a matter of degree—some people like me, I can close my eyes, I can tell you what the apple looks like, I can tell you what colour it is, I can tell you where the shading is. Someone like my wife doesn't see the apple. She can tell you what an apple is. It's really interesting because she has a degree in architecture, which is known as a very visual field. But the thing you discover about aphantasia is, it doesn't keep people from becoming artists. In fact, it's the opposite. Someone like Ed Catmull, who co-founded Pixar, writes about it in his book, and so many of the great animators at Pixar are actually aphantasics. The reason is that they learned that they had to draw in order to see things. When you don't have a picture in your head of what you want something to look like, things appear in the drawing, and you find things that you couldn't even picture. A lot of writers actually are aphantasics. John Green discovered recently that he has aphantasia. It turns out that it's a superpower for writers, because if you don't have a picture in your head, then you don't have to translate that picture into words. A lot of writers talk about thinking in radio, like they have a constant narrator. My wife—she's probably going to kill me for talking about her this much—when she describes it to me, she's like, “Oh, it's like a radio in my head. I'm constantly hearing a voice, and it's a narrator.” I was like, “Holy shit, that would be really helpful to me.” I don't have anything like that in my head. I read Mrs Dalloway for the first time, and I gave it to her and I said, “You've got to read this book. I think this must be what it's like in your head.” And she said, “Oh my God, it is.” Part of the thing that I took away from that experience—this is a long-winded way of getting here—is that I take a lot of inspiration from people with this condition. Most of the people I know in the arts or the creative fields, they set out with this grand vision, and then they start working on the thing and it's nothing like what they had in their head, and they get really depressed: “This isn't what I had in mind.” Whereas if you set out without a picture in your head, and you just start manipulating things and you see what appears, that's more of the comic mode I was talking about earlier. What would happen if we just sat down with our materials and we started playing and we saw what appeared on the page? What if we started typing and saw what appeared, and then we played with that? That's the kind of joy. That's more like how kids operate. Kids are better at that. They're better at reacting to what's actually in front of them, instead of having these grandiose visions about what they're trying to achieve. Jo: Just coming back on the longevity of a creative career. Your books are very distinctive. You have a very distinctive visual style, your handwriting and the way the books are done. I wondered if another part of the ennui, perhaps, or the draining of the later career is that we get trapped into doing something that feels like it looks the same. Or we have a voice, and we're happy in that voice, but sometimes we want to do something completely different. For authors, we have different names. I write under two different names, and that helps. But equally— How do you define author voice, and do you ever feel like doing something completely different to your normal style? Austin: Style, in a lot of ways, is self-plagiarism. Style is the repeated things that we notice in people's work. Hitchcock talked about this in films. Wes Anderson is someone like that—Wes Anderson has a style. I'm sure that he gets really sick of it too sometimes, but you also can't help it in some ways. I thought a lot about this because people worry about style so much. A lot of the time, what we call style is what Adrian Tomine one time said: “Style is just the distance between what's in my head and what comes out of my hand.” I really like that definition. With this book, I was trying to think, “Okay, if I do another book in this series, how can I push things a little bit?” And then I was reading this article about Taco Bell. You guys have Taco Bell over there, don't you? Do you have Taco Bell? Jo: No. Austin: So Taco Bell, for people who don't know, is this American Mexican chain, and they have tacos and burritos and stuff like that. They're well known for making these really insane… it's so American, this company. They make a taco with a Doritos as a shell. Doritos are crisps, I guess. Jo: Yes, we have Doritos. Austin: Okay. I spent time in England, I just don't remember if I ate Doritos when I was in England. Anyway, I was reading this article about Taco Bell. It was really funny. They have an innovation kitchen at Taco Bell, and they have a rule about new products. The rule is called the distinctiveness rule, and the rule is: you can change the flavour or you can change the taste, or you can change the form, but you can't change both at the same time. I got really obsessed with this concept because I thought, “Well, this could be kind of interesting.” If you're someone who's had success and you're known for something, this presents an interesting thing. You could do a complete break and do something completely new, or you could try the distinctiveness rule. Okay, well, what if I play with this idea of taste versus form? What if I change the taste and keep the form? So the idea for Don't Call It Art was, what if I do another one of these books, but the taste is more like if my kids made it? It had the texture of kids' art, it had lots of scribbles in it, it was loose and messy. That was kind of the idea. The actual book ended up being more like the other books. It ended up looking like an Austin Kleon book, because I just can't help that. The thing you said about having multiple names that you write under, that's kind of what I do with the newsletter. I think of the newsletter as very different from the books. The newsletter is this twice-weekly thing where I can be a little bit more of myself. In the books, I'm this very helpful, happy version of myself. It's me, but it's me on my best day. I'm really helpful and interesting for you. The newsletter is still a highlight reel in a sense, but it's a little bit more of my weird everything-I'm-into. It's more of the unclipped version of me. The newsletter becomes a place where I can do a lot of the weird stuff that's much different from the books. I have these little projects going all the time. Sometimes I'll make a bunch of prints and put them online. Sometimes I'll make a bunch of zines on a topic I haven't covered in the book. Sometimes I'll do a mixtape. As someone who's interested in a lot of different forms and genres and just different modes of output, having something like a newsletter has been really creatively fruitful for me. It's kept me from getting too bottomed out with the books because the books do a certain thing for the reader, and as much as I'd love to do a book that was radically different, I also think I've been given a real gift with the form of my books, in that I kind of own the way that they feel and look. There aren't a lot of books that look like those books and feel like those books, and so I like playing with that form. It would be hard to get rid of it now. The pseudonym for me is kind of like the newsletter in a sense. The newsletter is a little bit more of where I get to be wild and wacky. Then the books are a little bit more of a chiselled thing. Jo: The books are perfect examples of the form, as you say, but it's interesting about the newsletter. You mentioned at the beginning that we can be drained by the admin around the work. For many people listening, a newsletter becomes admin. So how does the newsletter fit into your business? The books are traditionally published, they're very professional. How do you have your independent side, and how does all of that work together in your business? Austin: Thank you for asking that question. I run the whole show at the newsletter. The newsletter is just me, and then my wife edits it, and no one else is involved. I don't have an assistant. I don't have a team. It is just me, and that's why I love it. I control everything. I pick who gets in there. I pick everything. I love that. I grew up watching David Letterman over here, and Letterman had a nightly show, and I always thought that was killer. I thought, “Man, what a fun job. You have a show every night where you have a new guest, and you have all these wacky things going on.” It was like a variety show. I always thought that would be really fun, so the newsletter is my version of that. I started the newsletter in 2013, and it was just a Friday newsletter. It quickly became a list of 10 things I thought were worth sharing. I had a friend, Hugh MacLeod, who was like, “Hey, I have a newsletter. It's bigger than any conference you've ever gone to.” He was talking about South by Southwest here in Austin. He's like, “I have a newsletter now, and it's bigger than South by Southwest.” Jo: Oh, I remember him. Austin: He would say, “Every time I have a new print, I put it out, and there's a button, and then they buy it.” He was like, “You've got to get it. This newsletter thing is killer.” This was in 2011 or something. Jo: Yes, I still have his books. Blogging in Your Underwear or something. Austin: Totally. So Hugh's a whole different story, but I was just like, “Oh, I should really get a newsletter.” Letterman always had a top 10 list on his show. I just always thought a 10 list was really fun. And of course the books are lists of 10 too. So it just worked to have a weekly list of 10. It felt good, and it felt like an infinitely repeatable format. What I'm looking for as a creative person is an infinitely repeatable format that can go on and on and on and be new every time. So the list of 10 is something that people know the form of. It goes back to the Taco Bell thing. They know the form, but they're not sure what's going to go inside. They know it's going to be a burrito, but they don't know what's going to be in the burrito, and that's the exciting part. The newsletter, business-wise, was always a marketing cost for about the first eight years of its existence. I paid MailChimp to send it out. Then in about 2021, when I hadn't done a book for a while, my agent said, “You know, you should really think about doing a paid tier of your newsletter.” And this is to his credit, because he doesn't make anything off the newsletter. He said, “There's this thing called Substack now that makes that really easy.” So we moved to Substack in 2021 in October, and I started doing a Tuesday edition of the newsletter that was just for paid people. That grew enough that it's gone from a marketing cost to something that's almost—it's not quite as much as I make on my books, but it's close. And to be candid, my books sell pretty well. So suddenly the newsletter has become this really healthy income stream. The newsletter to me is actually the day job now. The newsletter is what really keeps the lights on. It's also the perfect mix. It's the day job, it's the thing that keeps income coming in on a regular basis, but it's also the thing I like to do the most. I'm not like a traditional writer who likes to just get lost in their book and take years and years and go away. I'm someone who loves to be doing a lot of different things. The newsletter is a perfect format for me. I'm talking myself into not quitting, actually. It's funny. It's gone from this thing that was a marketing cost to now it's a significant part of our income. That journey—such a bad word, journey—that trip has been very interesting. It's been really cool. But I'm also just lucky. I've been really lucky, and I think part of my thing is, I'm always just trying not to squander my luck. Jo: Well, the book is fantastic, and I know people are going to love it. And the newsletter, of course. So tell us— Where can people find you and your books and newsletter online? Austin: The easiest thing to do is to just go to AustinKleon.com, and that has links to everything—the books, the newsletter. I do actually keep an old-school blog still. I'm one of the few people that still maintains their blog and keeps it up to date. I'm hedging my bets because I think in the end everything will come back to a self-hosted website. I think in the end everyone's going to just go back to their little websites, or at least I hope so. Jo: Well, that was great, Austin. Thanks so much. Austin: Oh, thank you. The post Don't Call It Art: Rediscovering Creative Joy With Austin Kleon first appeared on The Creative Penn.