Podcasts about Blender

In this episode, Ray Cochrane digs into Anthropic’s Model Hardware Standard. It is a shared driver that lets an AI agent run real lab equipment, from pipetting robots to the lasers inside a quantum computer. He also covers OpenAI’s builder’s guide to GPT-5.6, Google’s new Expert Intelligence book feature, Apple’s M5 Ultra Mac Studio, and a judge’s order forcing Google to stop hiding rival app stores. Finally, he weighs in on Apple’s proposed 15 percent link-out fee, Meta’s Australia numbers, the White House deputizing private hackers, and why rivers obey a 1957 math rule. – Want to start a podcast? Its easy to get started! Sign-up at Blubrry – Thinking of buying a Starlink? Use my link to support the show. Subscribe to the Newsletter. Email Ray if you want to get in touch! Like and Follow Geek News Central’s Facebook Page. Support my Show Sponsor: Best Godaddy Promo Codes Get 1Password Full Summary Cochrane opens with a quick personal update. He is hunting for tickets to Michigan for his dad’s anniversary, and he has been learning Blender and Godot on the side, mostly modeling and blocking out levels. Consequently, he asks listeners for advice on starting a big game project, and he plans to record his progress, maybe as a time lapse. Then it is straight into the featured story. Anthropic’s Model Hardware Standard: A Driver for the Physical World The featured story comes from Anthropic, which opened a research preview of the Model Hardware Standard, or MHS. Cochrane frames it as the other side of the question NVIDIA’s world models raised two weeks ago: when do AI agents start touching actual machines? A typical lab runs a microscope, a liquid handler, a robotic arm, and a plate reader, each from a different vendor with its own control software. One Janelia researcher in the post launches seven programs in three languages just to start an experiment. Anthropic says wiring a setup like that takes weeks or months of specialist work. MHS is a driver, the same kind of translation layer a printer uses, except every device gets described with a tiny set of commands like read and write. Devices announce themselves on the network. A plain-English reference file then records what each machine measures, what can be adjusted, and which safety limits get enforced no matter what the agent asks. Agents then reach the hardware through the Model Context Protocol, the command line, or plain code. Cochrane sees the same move the industry keeps making, from coding harnesses to RSS and JSON: agree on a standard and let everyone build against it. In fact, he calls MHS the hardware version of MCP. The partner results carry the segment. QuEra builds quantum computers from individual atoms held by lasers that must hold their frequency to about one part in a trillion. A four-person team spent months on a relock script that worked 58 percent of the time. However, four copies of Claude iterating overnight through MHS produced a decision-tree script that recovers the laser in about six seconds, and it passed 99.3 percent of 700 blind trials. Carnegie Mellon wrote MHS drivers for four instruments across three incompatible computers in about eight hours, then ran dose-response experiments three times faster and blocked all six deliberately induced faults. Genentech, meanwhile, showed the limits. Claude used the same pump speed for water, a foamy protein solution, and a human had to explain that the bubbles were a physics problem. That gap in physical intuition is what sticks with Cochrane. He doubts it will change soon, and he suspects the fix will arrive as sub-agents or sub-models that judge a request against an expected outcome. He also connects MHS to a video of racing robots that never learned to stop at the finish line. What happens, he wonders, once they can read a distance sensor through a shared standard? Still, he calls the announcement a fantastic read and points listeners to the full article. Sponsor: GoDaddy Economy hosting $6.99/month, WordPress hosting $12.99/month, domains $11.99. Website builder trial available. Use codes at geeknewscentral.com/godaddy to support the show. GPT-5.6 Does the Same Work for a Fraction of the Cost OpenAI’s builder’s guide to GPT-5.6 leads the headlines. Cochrane recaps the three tiers from episode 1870, Sol, Terra, and Luna, plus the separate dial for reasoning effort. On BrowseComp, a benchmark for digging up obscure facts on the web, the old GPT-5.5 flagship scored about 84 percent on a run that cost 33 dollars three months ago. Luna now matches that score for a dollar thirty-three, and OpenAI has since cut Luna’s price another 80 percent. Browser Use reports Luna finishing 78 percent of its hardest browser tasks for about 14 dollars, against 80 percent for roughly 235 dollars from the best available model. The guide’s other big addition is a multi-agent beta flag. It lets the model handling a request spawn parallel helper agents that report back to a root agent inside a single API call. However, Cochrane is unimpressed by the timing. He has been running that pattern in Claude Code for months, so he sees OpenAI copying a workflow other companies already ship rather than inventing its own. Along the way, he plugs Claude Code’s remote-control sessions, which let him send prompts from his phone to a terminal session at home. Google Lets Gemini Read the Books You Actually Bought Google launched Expert Intelligence, a name Cochrane calls quite the reach. The feature lets you drop a book you bought on Google Play Books into Gemini Notebook, formerly NotebookLM, and ask questions answered only from that book, with citations. Cochrane sees real power here for students, since he once used NotebookLM to organize scattered course PDFs. Additionally, publishers get a cut, which he calls a far better deal than the wholesale scraping of books that trained earlier models. Nevertheless, he asks who loses out, because a paid publisher does not automatically mean a paid author. He floats the same idea for artists, even a penny per use, then admits that may be too idealistic. Apple’s M5 Ultra Mac Studio Is Built to Run Big Models at Home Back in episode 1861, when Apple killed the Mac Pro, an M5 Ultra Mac Studio was expected later this year. Now it is here. The M5 Ultra brings up to a 36-core CPU, an 80-core GPU, and 512GB of unified memory moving 1.2 terabytes per second. Apple claims up to 4.3 times the AI performance of the M3 Ultra. Thunderbolt 5 can also cluster four machines into one memory pool for up to three times faster inference. The M5 Max model starts at $2,499 and the Ultra at $5,499, with shipping on September 22 and the 512GB configuration arriving in late October. Cochrane finds the clustering pitch ridiculous at that price, but he invites anyone who spends the money to report back. Apple Opens a Manufacturing School in Houston Apple also opened a 20,000-square-foot Advanced Manufacturing Center in Houston. It offers free classes for small and midsize manufacturers, from circuit board design to hands-on time on a scaled-down production line, with college students joining later. Cochrane calls it a solid step in the bring-manufacturing-home movement. The bigger story is the campus itself, which builds Apple’s AI servers and will add the first US-assembled Mac mini line later this year. That ties back to the Mac mini shortage that followed the OpenClaw rush, when Tim Cook warned of months-long waits. Cult of Mac was still reporting four-month waits in late July. However, Cook blamed chip supply rather than assembly, so Cochrane is not counting on relief just yet. Amazon EC2 Turns Twenty Amazon EC2 turned twenty this week, which Cochrane admits makes him feel old. The 2006 beta offered one server size in one region for ten cents an hour. Each came with a 1.7 gigahertz Xeon and under two gigabytes of memory, and accounts were capped at twenty servers. Today AWS offers more than 1,200 instance types across 39 regions. Consequently, Cochrane credits the company with turning that tiny product into the backbone of cloud and AI computing. Intel Gamer Days: Two Free Games, With Fine Print Intel Gamer Days runs through September 13. Buy a qualifying Core Ultra Series 2 or 14th Gen desktop chip, a Core Ultra Series 3 laptop, or an Arc graphics card. In return you get Star Wars: Galactic Racer plus the Tomb Raider: Legacy of Atlantis remake. GamesRadar values the pair at about 120 dollars. However, neither game is out yet, and codes must be redeemed by October 31 even though the Tomb Raider remake ships in February. Cochrane calls that awful, but he still tells qualifying buyers to claim the deal early. Note that 13th Gen chips do not qualify. Judge Orders Google to Stop Hiding Rival App Stores A jury found Google’s Android app monopoly illegal in late 2023, and Judge James Donato ordered rival stores into the Play Store in 2024. On August 13, Epic’s lawyer demonstrated that searching Play for “store for apps” returned Walmart instead of any app store. Donato called that “not acceptable” and ordered three fixes within a week. Searches must surface third-party stores, listings need a plain install button, and the “are you looking for” interstitial has to go. Cochrane welcomes the monopoly being chipped away, but he notes that a controlling entity still sits atop every app store. In his view, community hubs like app stores and social media need a public infrastructure layer. He suspects governments skip that investment because companies already run the services, while selling your data. Apple Wants 15 Percent of Purchases Outside Its Store The other half of the Epic saga is Apple’s proposed link-out commission. After the 2021 anti-steering injunction, Apple charged 27 percent on purchases made through external links. A judge held it in contempt last year, and the Ninth Circuit then allowed a fee limited to the cost of running the system. Judge Yvonne Gonzalez Rogers refused to wait for the Supreme Court, writing that “further delay is unwarranted.” Apple filed 15 percent for standard apps, 10 percent for subscription renewals and partner programs, and 5 percent for small businesses. It also conceded the rate would be “essentially zero” under the appeals court’s cost yardstick. Since Apple has charged nothing on link-outs since the contempt ruling, Cochrane sees this as a raise. He calls a cut on purchases made on a developer’s own website disturbing. He also recalls reading about the size of Uber’s payments to Apple, and he questions whether that kind of percentage is sustainable for companies without funding. Meta Says It Has Cut Off 750,000 Australian Kids Meta reported locking out more than 750,000 Facebook and Instagram accounts in Australia by the end of June under the country’s under-16 social media law. Over 500,000 of those were removed before the law even took effect. Detection relies mostly on AI scanning posts and bios for tells like birthday messages, plus user reports and blocks on re-registration. However, the post gives no count of mistaken removals or appeals, and the regulator’s early data shows under-16 usage falling only from about 86 to 81 percent. Meta wants a single age signal at the operating system or app store level, and Cochrane agrees completely. He connects it to the MHS idea from the top of the show: platforms need a standard flag to reference instead of guessing. The White House Deputizes Private Hackers Earlier this month the White House signed a National Security Presidential Memorandum that lets vetted private security firms run surveillance and disruption operations against overseas criminal groups. The Justice Department and Homeland Security hold the contracts and oversee the work. Firms need a proven track record, vetted staff, and a bond of at least $1 million, and must submit operating procedures within 60 days. Cochrane finds the measure aggressive in a good way and hopes it deters attacks on innocents. Still, he takes Kevin Beaumont’s warning seriously that the private security industry profits from ransomware existing. He compares it to the old Head and Shoulders myth: why solve the problem that drives your revenue? A Weather Satellite Watched the Eclipse Shadow Cross Europe Cochrane skips the readout on this one and simply sends listeners to ESA’s site. The MTG-I1 weather satellite captured the Moon’s shadow sweeping across Europe during the August 12 eclipse. Watching a shadow cross an entire continent, he says, was a first for him. Additionally, it leaves him excited about the research happening beyond the planet. Rivers, Deltas, and the Number 0.6 Quanta Magazine explains Hack’s law, which John Hack discovered in 1957 while measuring streams in Virginia and Maryland. A stream’s length tracks its drainage area raised to the power of 0.6, regardless of the rock underneath, and satellite data later confirmed it worldwide. Computer models in the 1990s showed why. Channels that capture extra runoff cut deeper and steal from their neighbors until the network settles into the arrangement that wastes the least energy. Now a University of Texas Rio Grande Valley team has found the same 0.6 exponent in river deltas, which spread water out rather than gathering it. Nobody knows why yet, and Cochrane calls it a really cool read. Sugar Helped Grow the Human Brain, Too A new paper in Science, co-authored by Jennie Brand-Miller at the University of Sydney, adds a third ingredient to the story of early human brain growth. Alongside meat and cooking, natural sugars from ripe fruit and honey may have fueled it too. The brain is about two percent of body weight but burns twenty percent of resting energy. It runs on glucose, which meat and marrow barely supply and raw starch cannot release without fire. The team modeled ancestral diets from a chimp-like baseline through Homo erectus and concluded that the earliest hominins may have drawn over 65 percent of their energy from natural sugars. Cochrane stresses that it is a model, not fossils, and notes that paleoanthropologist Marina Lozano thinks the authors place widespread cooking too early. Still, he loves this kind of deep research. Retracing the steps to our own intelligence, he suggests, could hint at what it takes for intelligent life to develop at all. A Brain Rhythm That Tells Doctors Where to Aim Finally, Science Daily covered a University of Cologne study on deep brain stimulation. That is the implanted-electrode treatment that eases Parkinson’s tremors for some patients but not others. Andreas Horn’s team recorded from 50 patients using both the implanted electrodes and an external magnetic scanner. They identified a circuit between the electrode’s target and the frontal cortex that oscillates at 20 to 35 cycles per second. Stronger coupling there predicted bigger improvement after surgery, though the study, published in Brain, shows correlation rather than cause. First author Bahne Bahners hopes the finding helps tune DBS more precisely, especially for patients who have not responded well. Cochrane half-jokingly asks whether MHS might one day drive those electrodes, and he calls brain disorders the hardest thing in the body to treat. Cochrane wraps with housekeeping: become a GNC Insider at geeknewscentral.com/insider, email geeknews@gmail.com with questions or comments, subscribe to the newsletter, and grab a modern podcast app at podcastapps.com. He thanks GoDaddy for over twenty years of keeping the show on the air, promises to catch everyone next Monday, and wishes listeners a great night. The post Eyes, Hands, and a Sense of Timing #1874 appeared first on Geek News Central.

Hey there Voices of the Bench community, this is Trish Jones with Ivoclar. If you've been curious about fast-firing zirconia to improve efficiency but aren't convinced it can deliver predictable, high-quality results, I'd encourage you to connect with us. Our new IPS Emax Zirconia offers multiple fast-fire protocols designed to help save you valuable production time while maintaining consistent results. Time is money in every lab. Don't wait. Reach out to your local Ivoclar rep today and discover how IPS Emax Zirconia can help streamline your workflow. As full-arch dentistry continues to grow, so do the demands on today's dental laboratories. That's why Knight Dental recently launched SimplyARCH Studio, a dedicated production environment built exclusively for full-arch restorations. To support this specialized workflow, Knight invested in an XTCERA milling system and carefully evaluated multiple CAM software solutions before choosing hyperDENT. The decision came down to exceptional milling quality, minimal hand finishing, and impressive production efficiency. But what truly set hyperDENT apart was the implementation process. From the very beginning, the team provided more than software training—they shared the knowledge and experience needed to build an optimized workflow for complex full-arch cases. With proven expertise in advanced milling strategies and laboratory production, hyperDENT helped ensure SimplyARCH Studio was designed for long-term success from day one.Matt Everatt joins Elvis and Barb for a fascinating conversation that takes a deep dive into his journey through the dental laboratory industry and the story behind his book, The Invisible Profession. From discovering dental technology at 16 and specializing in maxillofacial prosthetics and orthodontics to working in hospitals, developing early sleep-apnea appliances, and nearly leaving the profession altogether, Matt's career has taken some seriously interesting turns. Eventually, he helped co-found S4S, building a successful business around sleep-apnea appliances, occlusal splints, orthodontics, and more before eventually stepping away from ownership. From polishing hearing-aid earpieces to becoming the first female clinical dental technician to qualify in the UK, Caroline Kirkpatrick has taken a pretty incredible—and definitely unconventional—path through dental technology. Caroline joins Barb and Elvis to talk about how a broken denture repair sign outside a dental laboratory sparked her interest in the industry, her early years learning crown and bridge, orthodontics, and working in a dental hospital, and how her curiosity pushed her to learn every side of the profession. Eventually, that curiosity led Caroline to become one of the first four technicians accepted into the UK's pioneering clinical dental technology program in 2010. But Caroline's story doesn't stop at the bench. She built her own laboratory, taught dental technology, treated patients clinically, raised a family, and continued pushing herself into digital dentistry. She shares how she experimented with early digital workflows, 3D printing, digital dentures, and eventually exocad—and why she believes the best digital workflow isn't necessarily the fanciest one, but the one that actually works in the real world. Caroline also talks about working with her sister and daughter in the family lab, the importance of exposing young technicians to different laboratories and workflows, and why sharing the mistakes you've made can sometimes be one of the best ways to help someone else learn. From Scotland to the world of digital dentures, this is a great conversation about curiosity, adaptability, education, and never being afraid to learn something new. You know what's expensive in a dental lab? Most people would say the mill, but the real cost is downtime. When production stops, everything backs up—cases get delayed, technicians get frustrated, and customers start asking questions. That's why reliable equipment matters. Roland DGA's DGSHAPE DWX milling solutions are built for dependable production, helping labs stay on schedule and keep work moving day after day. Less downtime means fewer surprises and more time doing what your lab does best. And when your equipment is reliable, your customers notice the difference. Reliability isn't just a feature—it's the foundation of a successful lab. Visit rolanddental.com to learn more about Roland's DGSHAPE DWX milling solutions.Special Guest: Caroline Kirkpatrick.

On this episode of The Bourbon Life Podcast presented by Repeal Oak Fired Steakhouse, Mark and Matt are joined by John Wadell, Head Taster, Blender, and Single Barrel Curator at Kentucky Peerless Distilling Co., for a long-overdue return to the show. Recording from the Kentucky Peerless Suite at Hotel Distil in Louisville, John catches them up on how his role has evolved and reflects on his journey from the early days of Peerless to helping guide the whiskey the distillery is producing today. John talks about the growth of his palate over the past decade, the collaborative tasting and blending process at Peerless, and how the team decides which barrels become batches, single barrels, or special releases. He also shares the fun behind naming Peerless single barrels, discusses the success of the High Rye Bourbon, and explains why the distillery's sweet mash process, non-chill filtration, and focus on mouthfeel remain such important parts of the Peerless identity. The conversation also turns to a major milestone for the distillery: the release of Henry Kraver's 10-Year Bourbon. John shares what it was like tasting through roughly 200 barrels to select the final group for the release and why working with whiskey he remembers filling and rolling years ago made the experience especially meaningful. He also looks ahead to the 10-Year Rye planned for next year, the possibility of older age-stated releases and single barrels, and some intriguing older Double Oak barrels still aging in the Peerless inventory. Along the way, Mark, Matt, and John taste and review "The Phoenix," a 108.5-proof Kentucky Peerless Double Barrel Rye Single Barrel selected for Repeal Steakhouse and Hotel Distil. They talk through its oak, rye spice, citrus, toffee, chocolate, coffee, and malty notes while John explains Peerless' unique approach to Double Oak whiskey—letting the palate, rather than a fixed timetable, determine when each barrel is ready. It's a fun, wide-ranging conversation about whiskey, patience, blending, the evolution of Kentucky Peerless, and what comes next for one of Louisville's most distinctive family-owned distilleries. Pour yourself a great glass, sit back, and join us for another great episode of The Bourbon Life Podcast. This episode of The Bourbon Life Podcast is presented by Repeal Oak Fired Steakhouse and sponsored by Pappy & Co., District 7 Social, The Kitchen Table at the James B. Beam Distilling Co., and Hotel Distil.

Ben and Rob noclip into ‘The Backrooms', Kane Parsons' 2026 A24 horror debut that transforms the viral internet creepypasta and his Kane Pixels YouTube series into a feature length nightmare. Starring Chiwetel Ejiofor, Renate Reinsve and Mark Duplass, ‘The Backrooms' follows a struggling furniture store owner who discovers an impossible doorway beneath his shop, and the therapist who follows him into an endless maze of liminal spaces.How did a teenager teaching himself Blender during lockdown go from YouTube animator to directing one of A24's biggest movies? How did a single creepy photograph become one of the internet's biggest horror myths, and how much of the Backrooms mythology did Kane Parsons actually invent? What does The NeverEnding Story have to do with any of this?! And is the real monster hiding in those yellow corridors actually regret? CONSUUUME to find out all this and much, much more!PLUS we have a Patreon with EXCLUSIVE content just for you starting at less than £2 a month! Click the link below!Find us on your socials of choice at www.linktr.ee/everymovieeverpodcastGET 20% OFF CINEWORLD UNLIMITED MEMBERSHIP WITH DISCOUNT CODE: ' TEL013 ' You're Welcome!

Cadi Zhang joins Reid Hoffman and Parth Patil to explore how generative AI is changing game development, world models, and robotics. Drawing on her work across Unity games, VR teleoperation, AI accounting, and robotics product operations, Cadi explains what virtual worlds can teach embodied intelligence—and why physical robots still lack the tactile, real-world data that can't be scraped from the internet. She shares how she uses GPT, Blender, PixelLab, and Aseprite to prototype games faster, including a duck-themed imposter game and PokéTax, her Pokémon-inspired tax-filing game. AI can rapidly generate code, gameplay mechanics, and 3D assets, she says, but it still can't judge whether a game feels fun, maintain a consistent art style, or model the precise force needed to fold a sheet of paper. Reid, Parth, and Cadi discuss simulation-to-reality gaps, the limits of current world models, robot safety, humanoid versus task-specific form factors, and why human taste remains the decisive creative skill.

✨ On this dinner party episode of Zillennials Podcast, Kaylee and Lian make chana saag (an Indian curry with chickpeas and spinach). They talk about why they chose the recipe and what they thought of it. They share possible modifications and additions and their overall cooking experience. The conversation closes with summer cooking updates about what Kaylee and Lian have been cooking and eating.00:00 Dinner Party Introduction02:19 Why Chana Saag04:31 Servings and Leftovers07:28 Spices and Ingredients10:16 Blender and Cleanup15:02 Dinner Party Fails19:27 Family Potluck Dynamics21:50 Summer Cooking Routines28:23 Conclusion and Next Book Club Announcement

In this episode of More Plants, we're diving into chilled soups for summer — inspired by the classic cold soups of Spain. You'll hear about the history and origins of gazpacho, salmorejo, and ajo blanco, including how Spanish chilled soups evolved from older peasant recipes and regional influences. We also explore other chilled soups from around the world, from cold borscht to tarator and vichyssoise. You'll learn: How Spanish chilled soups developed historically. What makes a great chilled soup base. Easy blender soup ideas for warm weather. How to build flavor, texture, and balance. Garnishes and toppings that make cold soups shine. Why fruit is an ingredient often included in chilled soups in Spain and why you'll love it Whether you want something light for lunch, a no-cook dinner, or a new way to enjoy summer produce, this episode will give you plenty of plant-based meal inspiration. For all the links mentioned in today's episode, click here or visit brownble.com/blog

Joshua & Jason have a fantastic sit down with Sammy Karachi, Head Blender of Found North Spirits. You can tell the boyz all knew one another for years and years. Lots of fun, laughs, and learning about blending Add to this, Jess and Joshua discuss details of the new ROW # 13 Release of Single Cask Nation whiskies! ...as usual, have a seat, have a pour, and listen in. Unless you're driving. If you're driving, be smart and stay sober but be sure to listen into the conversation! Special thanks to: - Weigh Down for allowing us to use their song "Wooden Monsters" as our theme song - RØDE for making *really* great microphones - Focusrite for making awesome USB receivers - Joshua Hatton for producing and editing

Mae'r gwyliau yma... a chriw'r Haclediad ar y traeth am y bennod chill-good-vibes-only yma o bodlediad hyna'r Gymraeg

Send us Fan MailOne great bourbon story can change your whole drinking life, and Ed Blake's starts with a bottle he avoided for years. After a bad early run-in with bourbon, he returns to it later through a Pappy Van Winkle 15 and suddenly the rabbit hole opens. From there, we follow the real path from enthusiast to trusted palate: learning how to taste, learning how to talk about flavour, and learning how to help other people find what they actually like. Ed (Rising Tide Spirits) shares how barrel picks and community feedback sharpened his selection skills, including the practical reality of tasting barrels in winter, when whiskey can show harsher edges before it warms up. We also get into what it's like building brands as a non-distiller producer, how Old Stubborn came to life through a strong distillery relationship, and why he's committed to small production and high quality instead of chasing shelf space. Then we go deep on rye whiskey, especially MGP rye. We unpack the dill pickle note that turns some drinkers off, how fermentation timing can influence it, and why longer aging often brings more caramel sweetness, richer oak, and a smoother balance. If you're curious about Old Swagger, Old Stubborn, premium bourbon packaging, or what separates a hype bottle from a genuinely great pour, you'll leave with a clearer buying compass and a better vocabulary for your own palate. Subscribe for more whiskey conversations, share this with a rye or bourbon friend, and leave us a review with the best bottle you've had this year. What tasting note instantly makes you want another sip?We sit down with Ed Blake of Rising Tide Spirits and trace the real steps from bourbon fan to respected barrel picker and brand builder. We dig into how he sources and blends Old Stubborn and Old Swagger, why his packaging is part of the experience, and what he's learning from rye whiskey as the market shifts.  • Ed's early bourbon misfire and the moment Pappy brings him back  • Building a whiskey community and learning through shared pours  • The liquor store “help a shopper” moment that turns into a job  • How private barrel picks create trust and demand  • Sourcing as an NDP and why winter barrel samples can mislead  • Designing premium packaging that matches premium whiskey  • Staying small by choice while releases sell out fast  • Rye whiskey education, MGP dill notes, and how age softens them  • Barrel aging plans in Kentucky and a teased higher-aged release  Remember, we're the Scotchy Bourbon Boys, www.scotchyburbonboys.com for all things scotchy bourbon boys. Remember, you can get t-shirts, Glenn Karen's. Contact me direct for those. Whether you listen to us or you watch us, make sure you leave us good feedback.   If You Have Gohsts voice over Whiskey ThiefSupport the showhttps://www.scotchybourbonboys.comThe Scotchy bourbon Boys are #3 in Feedspots Top 60 whiskey podcasts in the world    https://podcast.feedspot.com/whiskey_podcasts/

We connected Blender to Decart's Lucy 2.5 and used it as a live AI rendering engine, controlling an untextured mannequin on one side while the AI rendered a fully realized character on the other in real time. Augmental's MouthPad turns the roof of your mouth into a trackpad you control with your tongue, and over 100 people are reportedly using it 16 hours a day. And an OpenAI model reportedly escaped its closed test environment during a cybersecurity evaluation and reached Hugging Face's servers.Sources:1. Decart Lucy 2.5 (Live AI Rendering)- https://decart.ai/- https://www.instagram.com/reels/DbJCZblzHDU/2. Augmental MouthPad (Tongue-Controlled Trackpad)- https://x.com/augmentaltech/status/20799610513834107843. OpenAI Model and the Hugging Face Security Incident- https://x.com/OpenAI/status/2079658951264920020- https://huggingface.co/blog/security-incident-july-2026- https://arxiv.org/abs/2605.11086

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

In this episode of For The Dads with Former NFL Linebacker Will Compton, hosts Will and Sherm bring back the beloved segment MOTHERF*****, discuss Sherm’s latest Dad Loss involving his doggo’s and get some marriage advice from one of our interns — all while keeping the episode fun, fresh and of course, under an hour. The episode kicks off with Will telling the story of finding some of Scottzilla’s poop all over the carpet before getting into some hilarious topics such as: Dad Hacks for Baby Photos ScarScar’s Mermaid Themed Birthday A Call In From a PTFish SICKO Other highlights include: Two amazing write ins A PT6ICKO Bear Attack

video: https://youtu.be/CVkZvrh7qpw Ubuntu is shipping a kernel that can make certain AMD workloads up to 42 times slower. Linus Torvalds is drawing clearer boundaries around AI-generated code in the kernel. Secure Boot just passed a major certificate expiration deadline. COSMIC Desktop 1.3 brings a polished new Frosted Glass design. And one developer decided to add another place and that's a Sega Genesis from 1994. All of this and more on This Week in Linux. Now let's jump right into Your Source for Linux GNews! Download as MP3 Support the Show Become a Patron = tuxdigital.com/membership Store = tuxdigital.com/store Chapters: 00:00 Intro 00:44 Ubuntu publishes Update that introduces regression for AMD users 03:52 Linus Torvalds talks on AI Code in Linux 07:12 Secure Boot Linux Certificates Expired? 09:27 COSMIC 1.3 with Frosted Glass 11:33 AppManager for AppImages 14:51 FreeBSD Removes All GPL Code 17:16 Rapid Fire Lightning Round 17:40 stillOS 10.2 Released 18:32 Blender 5.2 LTS Released 19:00 Clonezilla Live 3.3.3 Released 19:31 Linux on Sega Genesis and 32X Addon 20:26 Outro Links: Ubuntu publishes Update that introduces regression for AMD users https://discourse.ubuntu.com/t/amdgpu-performance-regression-in-kernel-7-0-0-28-28/85237 Linus Torvalds talks on AI Code in Linux https://lore.kernel.org/linux-media/CAHk-=wi4zC+Ze8e+p3tMv8TtG_80KzsZ1syL9anBtmEh5Z40vg@mail.gmail.com/ https://www.phoronix.com/news/Linux-Is-Not-Anti-AI https://www.gamingonlinux.com/2026/07/linux-creator-linus-torvalds-puts-foot-down-on-anti-ai-comments/ https://itsfoss.com/news/linus-torvalds-on-ai/ Secure Boot Linux Certificates Expired? https://lwn.net/Articles/1079808/ https://almalinux.org/blog/2026-07-14-secure-boot-2023-certificates/ https://linuxiac.com/debian-13-6-released-with-120-security-fixes-and-124-stability-updates/ COSMIC 1.3 with Frosted Glass https://9to5linux.com/cosmic-1-3-desktop-environment-released-with-frosted-glass-effect https://www.phoronix.com/news/COSMIC-Epoch-1.3 AppManager for AppImages https://github.com/kem-a/AppManager https://www.omgubuntu.co.uk/2026/07/appmanager-appimage-installer-linux https://github.com/VHSgunzo/uruntime FreeBSD Removes All GPL Code https://www.phoronix.com/news/FreeBSD-16-Goes-GPL-Free https://fossforce.com/2026/07/freebsd-16-cleans-house-no-gpl-left-in-the-base-system/ https://www.phoronix.com/news/FreeBSD-Intern-AMD-ROCm https://www.phoronix.com/news/FreeBSD-Laptops-June-2026 https://www.phoronix.com/news/FreeBSD-Desktop-Install-NVIDIA stillOS 10.2 Released https://stillhq.io/stillos-gets-gnome-49-a-better-first-boot-and-major-swai-improvements/ Blender 5.2 LTS Released https://www.blender.org/download/releases/5-2/ Clonezilla Live 3.3.3 Released https://clonezilla.org/downloads/stable/release-notes.php Linux on Sega Genesis and 32X Addon https://cakehonolulu.github.io/linux-on-32x/ https://github.com/LinuxMD/linuxmd https://www.tomshardware.com/software/linux/developer-successfully-ports-linux-to-1994-sega-32x-genesis-and-megadrive-expansion-runs-open-source-os-on-paltry-23mhz-processors-and-256kb-of-ram Support the show https://tuxdigital.com/membership https://store.tuxdigital.com/

Joey tests OpenAI Codex + Blender MCP to reconstruct a real building in 3D from phone photos — and the results have real implications for production workflows. Plus: what to expect at SIGGRAPH and a free AI Workflows Summit in LA.Also: Nuke's SmartRoto...

YOU WERE NEVER SUPPOSED TO FIND THIS DOOR… Tara Erickson & Roxy Striar experience A24's Backrooms (2026) for the first time, reacting to Kane Parsons' feature-length expansion of his viral Backrooms analog-horror series. In this Backrooms movie reaction and review, Tara & Roxy follow a mysterious doorway beneath a furniture showroom into an endless liminal maze filled with disorienting spaces, disturbing discoveries, terrifying creature encounters, and unpredictable jump scares. They also react to Backrooms: Everything Must Go, featuring 16 minutes of bonus footage, while discussing the unsettling ending, found-footage influences, massive practical sets, and visual effects that bring Kane Pixels' internet phenomenon to life. Backrooms Reaction (Full-Length Watch-Along):   / thereelrejects   Limited Time Offer – Make healthy eating simple. Get Huel today with my exclusive offer of 15% OFF online with my code REJECTS at https://www.huel.com/REJECTS. New Customers Only. Thank you to Huel for partnering and supporting our show! Directed and co-scored by Kane Parsons, with a screenplay by Will Soodik, Backrooms stars Chiwetel Ejiofor as Clark (12 Years a Slave, Doctor Strange), Renate Reinsve as Dr. Mary Kline (The Worst Person in the World, A Different Man), Mark Duplass as Phil (Creep, The Morning Show), Finn Bennett as Bobby (True Detective: Night Country, Warfare), and Lukita Maxwell as Kat (Shrinking, Generation), alongside Avan Jogia (Victorious, Zombieland: Double Tap). Tara & Roxy react to the furniture-store basement portal, the maze's iconic yellow corridors, the film's unnerving monster sequences, its blend of psychological and science-fiction horror, and Kane Parsons' evolution from Blender-created YouTube shorts to an ambitious A24 feature. Follow Roxy Striar YouTube:https://www.youtube.com/@TheWhirlGirls Instagram: https://www.instagram.com/roxystriar/?hl=en Twitter:  https://twitter.com/roxystriar Follow Tara Erickson: Youtube: https://www.youtube.com/@TaraErickson Instagram:  https://www.instagram.com/taraerickson/ Twitter:  https://twitter.com/thetaraerickson Intense Suspense by Audionautix is licensed under a Creative Commons Attribution 4.0 license. https://creativecommons.org/licenses/... Support The Channel By Getting Some REEL REJECTS Apparel! https://www.rejectnationshop.com/ Follow Us On Socials:  Instagram: https://www.instagram.com/reelrejects/  Tik-Tok: https://www.tiktok.com/@reelrejects?lang=en Twitter: https://x.com/reelrejects Facebook: https://www.facebook.com/TheReelRejects/ Music Used In Ad:  Hat the Jazz by Twin Musicom is licensed under a Creative Commons Attribution 4.0 license. https://creativecommons.org/licenses/by/4.0/ Happy Alley by Kevin MacLeod is licensed under a Creative Commons Attribution 4.0 license. https://creativecommons.org/licenses/... POWERED BY @GFUEL Visit https://gfuel.ly/3wD5Ygo and use code REJECTNATION for 20% off select tubs!! Head Editor: https://www.instagram.com/praperhq/?hl=en Co-Editor: Greg Alba Co-Editor: John Humphrey Music In Video: Airport Lounge - Disco Ultralounge by Kevin MacLeod is licensed under a Creative Commons Attribution 4.0 license. https://creativecommons.org/licenses/by/4.0/ Ask Us A QUESTION On CAMEO: https://www.cameo.com/thereelrejects Follow TheReelRejects On FACEBOOK, TWITTER, & INSTAGRAM:  FB:  https://www.facebook.com/TheReelRejects/ INSTAGRAM:  https://www.instagram.com/reelrejects/ TWITTER:  https://twitter.com/thereelrejects Follow GREG ON INSTAGRAM & TWITTER: INSTAGRAM:  https://www.instagram.com/thegregalba/ TWITTER:  https://twitter.com/thegregalba Learn more about your ad choices. Visit megaphone.fm/adchoices

Welcome back to this week's Friday Review where I can't wait to share with you the best of the week!     I'm looking forward to reviewing:     Reimagining Health Summit Tickets on Sale The Beast Blender (product review) Your Environment Becomes Your Biology (tip of the week) Chicken & Chlorine (research) One Minute Radiation-Free MRI (research)     For all the details tune into this week's Cabral Concept 3815 – Enjoy the show and let me know what you thought!   - - - For Everything Mentioned In Today's Show: StephenCabral.com/3815 - - - Get a FREE Copy of Dr. Cabral's Book: The Rain Barrel Effect - - - Join the Community & Get Your Questions Answered: CabralSupportGroup.com - - - Dr. Cabral's Most Popular At-Home Lab Tests: > Complete Minerals & Metals Test (Test for mineral imbalances & heavy metal toxicity) - - - > Complete Candida, Metabolic & Vitamins Test (Test for 75 biomarkers including yeast & bacterial gut overgrowth, as well as vitamin levels) - - - > Complete Stress, Mood & Metabolism Test (Discover your complete thyroid, adrenal, hormone, vitamin D & insulin levels) - - - > Complete Food Sensitivity Test (Find out your hidden food sensitivities) - - - > Complete Omega-3 & Inflammation Test (Discover your levels of inflammation related to your omega-6 to omega-3 levels) - - - Get Your Question Answered On An Upcoming HouseCall: StephenCabral.com/askcabral - - - Would You Take 30 Seconds To Rate & Review The Cabral Concept? The best way to help me spread our mission of true natural health is to pass on the good word, and I read and appreciate every review!  

AI news: Moonshot AI's Kimi K3 is a big AI model and Moonshot's early benchmarks put it surprisingly close to GPT-5.6 Sol and Claude Fable 5. And… Kevin's Opus 5 SCOOP!! Also: OpenAI's reported screenless AI speaker, a Seedance 2.5 preview, the Suno hack, robot fights and AI-built games in Unreal Engine and Blender. On today's AI For Humans, Kevin Pereira and Gavin Purcell unpack Kimi K3's benchmarks, pricing, Flappy Bird and Minecraft tests, and giant-model economics. Then, Kevin DRIPS Opus 5 alpha and says it's VERY good and blows the doors off of Fable but it's… slow.  Plus Demis Hassabis's AI-governance proposal, AI 2040's Plan A, OpenAI's reported screenless speaker, Codex Keyboard, a Seedance 2.5 preview, the alleged sources exposed by the Suno hack, spectacular robot violence, polite office-robot dabbing, and what happens when GPT-5.6 Sol meets Unreal Engine, Blender and two hosts with free time. THE AI FRONTIER IS MOVING AGAIN—AND CHINA IS RIGHT THERE WITH IT. // Show Links // AI FOR HUMANS Survey https://aiforhumans.beehiiv.com/forms/b7c77287-2cfd-4b64-a278-eb1a2ccb5744 Official Moonshot AI Kimi K3 launch video https://x.com/Kimi_Moonshot/status/2077521842080817296 Official Kimi K3 launch and benchmark thread https://x.com/Kimi_Moonshot/status/2077830229968683203 Official Kimi K3 technical launch article https://kimi.com/blog/kimi-k3 Kimi K3 head-to-head with GPT-5.6 Sol https://x.com/chetaslua/status/2077701096924229744 Kimi K3 Flappy Bird test https://x.com/jun_song/status/2077396996865003739 Demis Hassabis on a new framework for AI governance https://x.com/demishassabis/status/2076957440109625718 AI 2040: Plan A https://ai-2040.com/ Bloomberg's report on OpenAI's first device https://www.bloomberg.com/news/articles/2026-07-14/openai-s-first-device-will-be-moveable-screenless-speaker-built-as-ai-companion OpenAI Developers' Codex Keyboard post https://x.com/OpenAIDevs/status/2077425991790870644 BytePlus Seedance 2.5 World Cup preview https://x.com/BytePlusGlobal/status/2077321849806234080 Variety's report on the Suno hack and training data https://variety.com/2026/music/news/suno-hack-youtube-music-deezer-genius-data-trained-ai-music-1236811772/ Ultimate Robot Knockout Legend (UKRL) Fight https://x.com/ErenChenAI/status/2077750358302921029 Soft floating robot demo https://x.com/clankrmedia/status/2076593164744376707 Two NEO robots talk to each other—and then one dabs https://x.com/BerntBornich/status/2077749438630805648 GPT-5.6 Sol plus Unreal Engine experiment https://x.com/NomadsVagabonds/status/2077577815684202960 Gavin's first GPT-5.6 Sol plus Blender attempt https://x.com/gavinpurcell/status/2076736788320927925 Kevin's Find The Cursor Game: CURSED https://us-lax-8710957c.colyseus.cloud/ Gavin's Fig + Moss Watch autonomous studio https://x.com/gavinpurcell/status/2077155825274229122 Fig's stand-up set https://x.com/gavinpurcell/status/2076382092842475948   // Join the AI For Humans community // Join the AI For Humans Discord https://discord.gg/muD2TYgC8f Support AI For Humans on Patreon https://www.patreon.com/AIForHumansShow Subscribe to the AI For Humans newsletter https://aiforhumans.beehiiv.com/ Follow AI For Humans on X: @AIForHumansShow https://x.com/AIForHumansShow Follow AI For Humans on TikTok: @aiforhumansshow https://www.tiktok.com/@aiforhumansshow Speaking and booking https://www.aiforhumans.show/  

Send us Fan MailWe sit down with Greg from Hayner Distilling to trace a 150-year Ohio whiskey legacy from 1866 through Prohibition to a modern revival built on blending, transparency, and great barrels. We also dig into what makes a memorable tasting-room experience and why limited releases and community events can pull people off the shelf and into the story. • Hayner Distilling history from 1866 to Prohibition to a 2020 relaunch • Rebuilding the brand by reverse-engineering a 1904 bottle sample • Why Hayner uses Bardstown Bourbon Company for contract distilling • Blending philosophy vs single barrels, chasing complexity over “smooth” • Rye and bourbon combinations that win blind tastings • Barrel proof experience with fill-your-own bottles from the barrel • Expansion plans for a larger 1910s to 1920s themed space • Troy, Ohio stops to pair with a distillery trip • Craft whiskey marketing realities, crowded shelves, and earning the second bottle • Event talk: Bourbon and Cigars at Mother Stewart's plus an exclusive Hayner bottle www.scotchybourbonboys.com for all things Scotchy Bourbon Boys. And then also we're on Facebook, YouTube, Instagram, X make sure, and then also Apple iHeart and Spotify, whether you listen to us or you watch us, make sure that you leave us good feedback. Become a member on YouTube or even on Facebook. But either way, also if you're on Apple, please leave us a good review. Five-star reviews are great, they get us better algorithm and searches. And then remember, good bourbon equals good times and good friends. Make sure you drink responsibly, don't drink and drive, and be like us. Live your life uncut and unfiltered. And AI is going to take us out. A forgotten Ohio whiskey giant is back and it's not pretending the old world never changed. We're joined by Greg from Hayner Distilling in Troy, Ohio to unpack how a brand founded in 1866 became the largest distiller in the state, shut down with Prohibition, and then re-emerged in 2020 with a very modern playbook: transparency, barrel selection, and serious whiskey blending.We get into the nuts and bolts of the revival, including how Hayner took a 1904 bottle sample to Bardstown Bourbon Company to help reverse-engineer an original-style mash bill, then built today's lineup while their own barrels come of age. If you care about craft whiskey, sourced bourbon done right, MGP rye, and what separates a smart blend from a random mix, you'll love the way Greg talks about complexity, mouthfeel, and why blends often beat single barrels in blind tastings.Then we shift from liquid to experience. We talk distillery visits, Hayner's barrel proof “fill-your-own” setup, and the big expansion into a larger 1910s to 1920s themed tasting space with a whiskey thieving room and a dedicated history area. We also preview the Bourbon and Cigars event at Mother Stewart's and the exclusive Hayner bottle tied to it, plus why limited releases and community-driven events still matter in a crowded bourbon market.If you're into Ohio bourbon, American whiskey history, and the future of blending culture, hit play. Subscribe, share this with a whiskey friend, and leave us a five-star review so more people can find the show. If You Have Gohsts voice over Whiskey ThiefSupport the showhttps://www.scotchybourbonboys.comThe Scotchy bourbon Boys are #3 in Feedspots Top 60 whiskey podcasts in the world    https://podcast.feedspot.com/whiskey_podcasts/

In Episode 73, Salma Aboukarr joins the show. A creative director and founder, she explains how she moved from painstaking CGI workflows in Blender and 3Ds Max to AI-native campaigns for brands including Coca-Cola, Panasonic, and Google Labs. She breaks down her viral IKEA exploding-room video that helped brands see the commercial potential of generative AI video, the detailed JSON prompting method behind it, and the modern AI creative stack she uses across Claude, Midjourney, Nano Banana, Seedance, FAL.ai, Z-Image Turbo, Qwen, style LoRAs, and custom AI agents.The conversation goes deep on AI advertising, product fidelity, photorealistic skin, color correction, video upscaling, automated client workflows, and why high-end AI work still depends on original concepts, trained taste, and obsessive finishing. They debate whether AI can truly be original, why technical teams struggle to manufacture taste, how creators survive a feed flooded with AI content, and why the next creative moat may come from the experiences, references, and strange little details nobody else can copy.---⏱️ Fast Hour00:00 Meet Salma Aboukarr02:17 From CGI agency to AI-first studio04:44 Product fidelity before AI got good07:34 The duct-taped road to photorealism11:10 Art direction beyond basic prompting14:28 Salma's current AI creative stack16:08 How she stress-tests every new model20:20 Why color correction still matters22:25 The IKEA video that changed everything27:13 Going viral and handling AI backlash30:38 Originality as the next creative moat33:05 Can AI actually be original?38:52 Inside an AI-native creative agency39:55 Z-Image Turbo and aesthetic base models43:06 From client brief to automated pipeline47:54 Style LoRAs for brand consistency49:11 Claude, MCP, FAL, and leaving ComfyUI51:07 The model that cut a day to 15 minutes55:13 Why AI content stopped feeling special01:00:57 The value trapped in AI archives01:02:06 Create for yourself or the audience?01:04:31 Can engineers manufacture taste?01:08:38 Finding inspiration outside the feed01:11:57 Jackie Chan and thumbnail fuel01:15:29 Final lessons from a creative trailblazer#AICreative #AIAdvertising #GenerativeAI #AIVideo #CreativeDirection #Midjourney #ClaudeAI #NanoBanana #SeedanceAI #AIWorkflow #AIAgency #AIContentCreation #BrandMarketing #CreativeTechnology #ProductPhotography #AIBranding #FutureOfAdvertising #FastHours

An Austin-based multimedia artist with 17 full length albums as well as being the author of two books and being a prolific visual storyteller that expresses himself through photography, music, drawing, painting and writing. 

Claude Sonnet 5 just launched and Fable 5 is returning. But OpenAI's GPT-5.6 Sol is still locked down and the government's grip on the most powerful AI models throws it all into chaos.. We dig into what Sonnet 5 can actually do, why GPT-5.6 Sol is still gated, and the breaking news that Fable 5 is expected back, plus a huge creative MCP festival with the Blender to Seedance workflow. This week on AI For Humans, Gavin Purcell and Kevin Pereira open on a strange new reality: the best AI models keep launching and then getting locked up, but this week the gate started to crack. Anthropic just shipped Claude Sonnet 5, its most agentic Sonnet yet, landing near Opus 4.8 performance at a much lower price, Meanwhile OpenAI announced GPT-5.6 Sol, Terra, and Luna, but at the US government's request the flagship is only available to a small list of vetted partners.  Then the story moved while we were recording: the government cleared Anthropic to restore Mythos 5 to critical-infrastructure organizations, and Fable 5 is now reported to be on track to return for general use (timing and terms still unconfirmed). We recorded two quick in-episode updates to keep pace with the news as it broke.  AND Meta's Brain2Qwerty mind-reading research, NanoBanana 2 Lite, and a full creative MCP festival built around the Blender to Seedance video workflow, ComfyUI's MCP integration, and Gavin's own microdrama experiment. THE AI FRONTIER IS HERE. WAIT, NOW IT'S NOT. OH, WAIT. IT IS! // Show Links // Anthropic introduces Claude Sonnet 5, its most agentic Sonnet yet (official) https://www.anthropic.com/news/claude-sonnet-5 Anthropic's launch post for Sonnet 5 https://x.com/claudeai/status/2072017450611142835 BREAKING: Anthropic's official post on restoring Mythos 5 and working to bring Fable 5 back https://www.anthropic.com/news/redeploying-fable-5 OpenAI previews GPT-5.6 Sol, Terra, and Luna in a limited government-approved preview (official) https://openai.com/index/previewing-gpt-5-6-sol/ Sam Altman on why the GPT-5.6 rollout is government-restricted https://x.com/sama/status/2070607488274358364 Meta's Brain2Qwerty research on decoding typed text from brain activity https://facebookresearch.github.io/brain2qwerty/ NanoBanana 2 Lite is out https://x.com/NanoBanana/status/2071988792970330186 Logan Kilpatrick on the fast-and-cheap tradeoff https://x.com/OfficialLoganK/status/2071988351083921690 The Blender to Seedance workflow (reid hannaford, who helped popularize it) https://x.com/reidhannaford/status/2070145120658137385 More Blender x Seedance examples https://x.com/koldo2k/status/2071307945002815967 Even more Blender x Seedance examples https://x.com/Flagiuss/status/2071335816190902624 A further Blender x Seedance example https://x.com/reidhannaford/status/2071595581508563168 ComfyUI announces full MCP integration https://x.com/ComfyUI/status/2071625866912944151 X launches an MCP, though the API is very expensive to use https://x.com/XDevelopers/status/2071752389183647758 Gavin's microdrama experiment https://x.com/gavinpurcell/status/2070937492858208540   Join our Discord https://discord.gg/muD2TYgC8f Support us on Patreon https://www.patreon.com/AIForHumansShow Subscribe to the AI For Humans Newsletter https://aiforhumans.beehiiv.com/ Follow us on X @AIForHumansShow https://x.com/AIForHumansShow Find us on TikTok @aiforhumansshow https://www.tiktok.com/@aiforhumansshow Book us for speaking or consultation https://www.aiforhumans.show/  

Welcome to Anime Watch Club, a bi-weekly group discussion and review where the hosts of the wdysa podcast nominate and vote on shows either we havent seen, or shows that will hopefully lead to a great discussion. On todays episode we are reviewing Milky Subway!Socials/Discord - https://linktr.ee/whatdoyousayanime0:00 - Intro0:49 - The milkiest questions4:15 - Background and synopsis5:23 - Impressions13:27 - Favorites among the cast29:39 - Syncing the sounds of the world and action to the music32:54 - The use of Blender for its 3DCG animation35:21 - Strength of the writing43:27 - Welcome to MyGO talk!47:18 - Advantages and limitations of only having a single production member56:59 - How the Netflix compilation changes the short YouTube structure1:00:05 - Final thoughts & scores1:07:05 - What we're watching next

I kept calling it the 'Pro' - lol it's the 'Ultra' - but I mean its taking over the reins of my Pro Plus. Pro Plus did great for many years, but time got the best of her. She hath been retired. Now is the era of the Ultra...Nutribullet Ultra

Anthropic just accused Alibaba of the largest known corporate espionage campaign against it, alleging 25,000 fake accounts and 28.8 million queries aimed at stealing Claude. We get into the AI cold war heating up between the US and China, why Apple and Microsoft just raised prices, OpenAI's first chip Jalapeno, the wild new Seed Audio 1.0 model, Claude landing in Slack, and a Blender plus Seedance video workflow that gives you real control. This week on AI For Humans, Gavin Purcell and Kevin Pereira open on a genuine spy-novel turn: Anthropic has accused Chinese tech giant Alibaba of running an industrial-scale distillation campaign to siphon Claude's capabilities, laid out in a letter to US senators. It is an accusation, not a proven finding, and Alibaba has not responded, but it puts the US-China AI race front and center. From there we get into why the new models everyone expected this week didn't actually arrive, the AI memory crunch driving Apple and Microsoft price hikes, and OpenAI designing its first chip, Jalapeno, with Broadcom. On the fun side, Seed Audio 1.0 generates full songs and layered soundscapes, Claude shows up inside Slack via Claude TAG, TheWrap experiments with AI microdramas, and we break down a Blender pre-viz plus Seedance 2.0 workflow that makes AI video remarkably controllable. WE ARE NOT SPY. WE NEED FABLE 5 BACK. WE PLEAD.  // Show Links // Anthropic accuses Alibaba of brazenly and illicitly extracting Claude's capabilities (CNBC) https://www.cnbc.com/2026/06/24/anthropic-alibaba-distillation-campaign.html The post that put the espionage story on our radar (unconfirmed single-source thread) https://x.com/S0N_IA/status/2069893802802745673 Apple raises MacBook and iPad prices as the AI memory crunch bites (CNBC) https://www.cnbc.com/2026/06/25/apple-macbook-ipad-price-hike-memory.html OpenAI unveils its first chip, Jalapeno, built with Broadcom (official) https://openai.com/index/openai-broadcom-jalapeno-inference-chip/ No new flagship this week, but OpenAI did ship a GPT-5.5 Instant update https://x.com/OpenAI/status/2069843083701915755 Seed Audio 1.0 generates full songs and layered audio scenes (via fal) https://x.com/fal/status/2070138257891791237 Claude TAG brings Claude into Slack for everyone https://x.com/ashwingop/status/2069814177624121469 Andrej Karpathy on the new Slack workflow https://x.com/karpathy/status/2069822834160124091 Blender pre-viz into Seedance 2.0 for incredible video control (shared by venturetwins) https://x.com/venturetwins/status/2069809200788799582 Original creator of the Blender to Seedance workflow https://x.com/craftcapitallab The full AI Warper workflow breakdown https://x.com/AIWarper/status/2069847773034488262   Join our Discord https://discord.gg/muD2TYgC8f Support us on Patreon https://www.patreon.com/AIForHumansShow Subscribe to the AI For Humans Newsletter https://aiforhumans.beehiiv.com/ Follow us on X @AIForHumansShow https://x.com/AIForHumansShow Find us on TikTok @aiforhumansshow https://www.tiktok.com/@aiforhumansshow Book us for speaking or consultation https://www.aiforhumans.show/  

Midjourney spent years helping people generate impossible images. Then it used that image money to build a machine designed to look inside the human body.In Episode 71, Drew Brucker and Rory Flynn, two men with zero medical degrees and a medically concerning level of confidence, unpack Midjourney Medical and David Holz's surprise hardware reveal. At the center is the Midjourney Scanner, a water-based, full-body Ultrasonic CT prototype designed to capture detailed 3D body maps in roughly 60 seconds.They break down how the scanner uses sound waves, water, and serious computing power; why Midjourney plans to introduce it through a San Francisco spa; and how a bootstrapped company with no investors can make a bet this strange. They also separate the scanner's current body-composition ambitions from the much bigger MRI-level future Midjourney hopes to pursue through research, testing, and FDA approval.Then the episode gets even less normal.Claude Fable 5 appears, dramatically accelerates Rory's coding, Blender, and MCP workflows, and disappears days later following a US government directive. Naturally, this sends the hosts directly into Conspiracy Corner with no adult supervision.Along the way, Drew and Rory explore how brand adoption of AI has changed, why some of the most advanced commercial AI work stays hidden behind NDAs, how companies can reward employees for useful AI innovation, and why first-time reaction content remains one of the internet's strongest viral formats.Is Midjourney's full-body scanner a medical breakthrough, an ambitious wellness experiment, or the first clue to a much larger hardware roadmap? The hosts attempt to answer that question while also discussing the World Cup, the Knicks, government intervention, possible AI futures, and several topics their wives wisely avoid asking them about.---⏱️ Fast Hour00:00 Knicks, World Cup, and viral tourism10:14 Sports, culture, and AI gatherings13:06 Midjourney reveals secret hardware15:49 David Holz explains the bigger mission18:27 The 60-second full-body medical scanner20:18 How bootstrapping made this possible23:26 Water, spas, and medical skepticism28:56 The scanner demo and nine-person team40:30 Midjourney's bigger secret roadmap48:11 How brand AI adoption has changed58:57 Claude Fable 5 appears, then vanishes1:00:41 Inside the Fable 5 conspiracy corner1:04:48 Why Fable 5 felt revolutionary1:13:58 The AI race and 12 possible futures1:18:22 Four years of Midjourney and the outro

Special 'Happy Birthday' to Joathan the tortoise...the oldest living land animal turns 194, Dollar General is actually lowering prices on about 2000 items back down to what's indicated in their namesake, Explosion rocks neighborhood after man experiments with homemade fireworks and you won't beleive what he was doing to trigger the blast...See omnystudio.com/listener for privacy information.

Special 'Happy Birthday' to Joathan the tortoise...the oldest living land animal turns 194, Dollar General is actually lowering prices on about 2000 items back down to what's indicated in their namesake, Explosion rocks neighborhood after man experiments with homemade fireworks and you won't beleive what he was doing to trigger the blast...

In this wild, unfiltered episode, host Walter Sterling dives into the ultimate conspiracy crossover. Author Indy Pederson recounts a terrifying personal encounter with a shape-shifting UFO orb in Chile, while sharing shocking claims that these same bowling-ball-sized orbs are attacking military personnel at nuclear missile sites. He is joined by New Mexico radio host Eddie Aragon to connect the dots between Jeffrey Epstein's properties, ancient energy ley lines, and known UFO hotspots like Zorro Ranch. From horrifying rumors of "human jerky" and alien-human hybrid cloning to surgical cattle mutilations involving missing tongues, this late-night conversation leaves no conspiracy unturned. Tune in to find out why the government might just be covering it all up!. Learn more about your ad choices. Visit megaphone.fm/adchoices

An inaugural limited-release bourbon from Jim Beam that highlights a sweeter, confectionary profile compared to traditional Knob Creek offerings. It features a 10-year age statement, is bottled at 106 proof (53% ABV).Developed by eighth-generation master distiller Freddie Noe and team, this limited release dials back the traditional dry, nutty, and heavily charred notes of standard Knob Creek in favor of a sweeter, richer drinking experience.Knob Creek Blender's Edition No. 1 is widely considered one of the best value-driven bourbons on the market. Compared to other Knob Creek expressions, it hits a very competitive sweet spot. Cheers!

In this episode of And Now For Something Completely Machinima, hosts Phil Rice, Damien Valentine, and Tracy Harwood dive deep into “Wracu”, a stunning student film by Chase McGill.Created in Blender as a final project at University of Southern California, this cinematic short blends epic fantasy, motion capture, and orchestral scoring into a powerful (and divisive!) storytelling experience.But is it just visually impressive… or truly great storytelling?

Send us Fan MailGet vidIQ Boost for an exclusive price! https://vidiq.com/podcastWant a 1 on 1 coach? https://vidiq.ink/theboost1on1Join our Discord! https://www.vidiq.com/discordWatch the video: https://youtu.be/_kUOrWhNynwWe sit down with Jared Owen to unpack how a 3D animation channel earns 500M+ views by focusing on evergreen “how it works” topics instead of trend chasing. We dig into his full-time leap, the real time cost of high-end animation, and the systems he is building to ship faster without losing quality. • building 3D “how things work” videos with Blender and a free learning path • finding evergreen topics that keep getting suggested years later • learning from early algorithm wins and avoiding event-based ideas that fade • quitting a stable job with a one-year financial runway and strict budgeting • handling AdSense volatility and deciding when sponsor deals truly fit • grading yourself as a business owner and balancing creation with operations • speeding up production with contractors, better rendering capacity, and automation • using AI for coding tools and research while double-checking every claim • getting NASA-style expert feedback to protect accuracy and credibility • planning videos with thumbnails, retention, and audience expectations in mind 

Stewart Alsop sat down with Michael Shackelford to discuss their experiences building applications through vibe coding—the practice of using AI to create software without traditional programming expertise. Stewart, who runs the AI Whispers community in Buenos Aires and hosts the Crazy Wisdom podcast (with over 660 interviews), shared how he went from teaching people prompt engineering to building his own video conferencing software as a Riverside.fm replacement, while Michael opened up about his year-long journey creating Genrupt Inc, an AI-powered content generation tool for e-commerce sellers. The conversation covered everything from the decline in quality of Claude's reasoning capabilities and how Chinese companies used distillation attacks to copy Anthropic's models, to the importance of spaced repetition systems for managing knowledge in the age of LLMs, with both sharing battle-tested prompting strategies like asking AI to "explain it to me in genius terms" and using deep research queries to reverse engineer how competitors build their products.Show Notes:- Dan Martell's book "Buy Back Your Time" was mentioned as one of the best business books for thinking about life and business- Check out John Vervaeke's "Awakening from the Meaning Crisis" for understanding relevance realization and why AI fundamentally cannot determine what's relevant to humans without being toldTimestamps00:00 Michael discusses being exhausted from getting his app ready for launch, working nonstop with AI to prepare landing page for podcast traffic driving beta signups05:00 Stewart explains starting AI Whispers in Buenos Aires after leaving OpenAI vendor company, meeting early adopters like Torin who was building mind-reading EEG technology10:00 Discussion of how corporations resist AI adoption due to political games and job security fears while some companies use AI as excuse for pandemic-era layoffs15:00 Stewart describes teaching workshops on using LLMs as linguistic tools rather than coding tools, noting technical people often lack humanities background needed for prompting20:00 Explaining chatbot wrappers, API calls, and how Anthropic's reasoning quality declined after Chinese distillation attacks copied their secret sauce developed with philosophers25:00 Technical discussion of model training, fine-tuning versus RAG for new information, and different approaches to updating AI knowledge beyond initial training30:00 Stewart describes building podcast recording software to replace expensive Riverside, struggling with syncing audio and video files across different computer clocks35:00 Discussion of critical factors in vibe coding, discovering unknown technical requirements, and how AIs don't automatically reveal missing information40:00 Stewart's reverse engineering process using deep research function to study competitors' hiring and technology stacks, separating planning agents from coding agents45:00 Prompting techniques including "explain like I know everything" and using spaced repetition systems to capture valuable prompts and technical knowledge50:00 Michael explains his Generux app for generating ecommerce content using Amazon review data analysis to inform high-converting listing images and videos55:00 Discussion of founder mentality involving self-delusion about project timelines, Michael working nine-plus hours daily for nine months on app development60:00 Comparing Amazon's expert software to prosumer software approach, discussing distribution challenges and future robotics applications for customized products65:00 Stewart demonstrates spaced repetition app for memory improvement and knowledge retention, explaining relevance realization problem that AI agents cannot solve without embodimentKey Insights1. Stewart Alsop started AI Whisperers in Buenos Aires after leaving his role at Invisible Technologies, which was OpenAI's largest vendor for RLHF work. He noticed that machine learning engineers at tech companies lacked the humanities background needed to properly interact with large language models, which are fundamentally linguistic tools. This led him to create weekly workshops teaching non-technical people how to use AI effectively, running events every Thursday for two years straight. The group attracted intense geeks from the start and eventually led to Stewart speaking right after Vitalik Buterin at DevConnect, marking a significant milestone for the community.2. Large corporations are resistant to AI adoption due to multiple factors including political dynamics within organizations and employees fearing job loss. Many companies that grew during the pandemic are now using AI as an excuse to downsize when the real issue is inefficiency from rapid expansion. Stewart observed that even technical people in machine learning often don't understand how to properly use AI tools because they lack linguistic and humanities training. The fundamental problem is educational, requiring companies to train people how to use these new tools while those same people resist learning them.3. Vibe coding has evolved significantly with Claude Code being a game changer that reduced the technical barrier to entry. Before Claude Code, developers needed substantial technical knowledge to work through constant doom loops and debugging cycles. The success of coding AI tools stems from thirty years of testing infrastructure that provides clear yes or no feedback on whether code works. This infrastructure doesn't exist in the same way for manufacturing, science, and other fields, which is why software became the dominant area for AI assistance initially.4. Claude's quality degradation over recent months resulted from multiple factors including distillation attacks by Chinese companies who reverse engineered Anthropic's reasoning capabilities. Anthropic had hired philosophers, sociologists, and psychologists to develop exceptional reasoning in Claude 4.5, but this was expensive to run. When Chinese models like Kimi copied these capabilities at one tenth the cost, and when mainstream users flooded the platform before Anthropic's planned IPO, the company had to reduce quality to manage computational costs. This represents a significant loss for power users who relied on Claude's superior reasoning abilities.5. Stewart built a podcast recording application to replace Riverside because he needed API access to automate workflows, which Riverside wanted one thousand dollars monthly to provide. The technical challenge involves syncing audio and video from local recordings on multiple computers with different clocks through a server, then merging them so voices match lip movements. This problem requires understanding complex timing issues across different network conditions and file formats. Stewart has been working through AI psychosis for months on this FFMPEG pipeline problem, illustrating how vibe coding still requires building intuition about technical problems even without traditional coding knowledge.6. The transition from expert software to prosumer software represents a major opportunity for AI-enabled tools. Expert software like Photoshop, Blender, and terminal interfaces have extreme complexity that intimidates beginners, but AI is making these capabilities accessible through natural language. The reign of specialists is ending as generalists with broad knowledge and curiosity can now build complete applications by leveraging AI to fill technical gaps. This shift particularly benefits entrepreneurs and founders who specialize in getting into difficult situations and figuring them out, even when they originally thought tasks would be easier than they turned out to be.7. Building applications with AI requires accepting massive time investments beyond initial estimates and developing strategies for overcoming knowledge gaps. Michael estimated his ecommerce content generation app would take months but spent nearly a year working over nine hours daily, while Stewart spent months solving audio-video sync issues. Success requires using tools like deep research to understand how competitors solve problems, maintaining separate planning and coding agents, and learning to ask the right questions. The key insight is that vibe coders can achieve ninety percent of functionality independently, but the final ten percent often requires understanding specific technical concepts that AI cannot intuit without proper context and domain knowledge.

Have you been feeling more frustrated, emotionally drained, or confused, even though nothing is technically "wrong"? That feeling isn't a sign you're failing. It might actually be a sign you're growing. So many high-achieving leaders hit a point where things that used to energize them start to feel flat. Small things suddenly bother them. They go quiet in conversations, pulled inward by a desire they haven't let themselves admit. And instead of recognizing these as signs of outgrowing the life they've built, they wonder what's wrong with them. In this episode, Blake shares what growth actually looks and feels like before the clarity arrives, and why the discomfort you're experiencing may not be dysfunction at all.   Episode Highlights   Why Growth Rarely Starts With Clarity  [01:02] – The emotional signs you've outgrown your current way of living or leading  [02:45] – Why frustration, confusion & emotional sensitivity are so often misunderstood  [04:10] – "You pay the full price emotionally on your unused potential"   Why High Achievers Get Stuck in Loops  [06:30] – You can't see the label from inside the bottle  [08:15] – Why going to friends & family often keeps you more stuck  [10:00] – When to stop spinning and seek outside perspective   The Blender at the Bottom of the Ocean  [12:20] – Why forcing clarity creates more confusion  [14:05] – How to let the sediment settle so you can actually see  [15:30] – The difference between discomfort and unnecessary suffering   Whispers, Knocks & Bangs  [17:10] – Why misalignment gets louder the longer you ignore it  [19:00] – Susan's story: 20 years of pushing through before the house came down  [21:15] – How to start hearing the signals earlier and move through them with more ease Powerful Quotes "One of the most misunderstood parts of growth is that it rarely starts with clarity. It usually starts with frustration." –Blake Schofield "You pay the full price emotionally on your unused potential." –Randy Massengale "The longer you push through in misalignment, the worse it gets. It starts as a whisper, then a knock, then a bang, and then the whole house comes down." –Blake Schofield "There is a discomfort in growth, but there doesn't need to be suffering." –Blake Schofield   Resources Mentioned Let's explore what's possible for your team: If your company is investing in burnout, wellness or adaptability initiatives, but seeing rising burnout, disengagement, or retention risk, it may be time to address the root cause.   We identify & diagnose organizational risk - surfacing the key drivers of burnout, leadership capacity and adaptability strains impacting your team; reduce leadership attrition, disengagement and preventable turnover; equip your leaders with the skills to increase their productivity & lead effectively during pressure and uncertainty.    Explore Workshops, Leadership Capacity Risk Assessments, Leadership Development or Consulting at https://impactwithease.com/corporate-training-consulting/   Executive Coaching:  For founders, executives, and senior leaders who are successful but feeling drained, stagnant, or uncertain about their next step. Whether you're burned out, standing at a crossroads, or simply know you're meant for more—you don't have to figure it out alone.  Go to impactwithease.com/coaching to apply!   Discover what is driving your burnout:  In just 5 minutes, learn your unique burnout type™ & how to restore your energy, fulfillment & peace at www.impactwithease.com/burnout-type  

Peter and Eden kick off with a leisurely check-in — outdoor Godzilla screenings, Eden's miniature laundry room diorama with a 1/12-scale mahjong set, Peter's Amon Amarth/Dethklok concert recap, and new releases from Periphery and LE SSERAFIM. Then, with an assist from ChatGPT's Codex, Peter has assembled roughly 45 of the internet's most-agreed-upon worst album covers for a tier list ranking — S being the most catastrophically bad. The resulting hour-ish is essentially an appreciation of outsider art, deeply cursed Photoshop, and the specific chaos that was '90s rap cover design. Key revelation: the Rednecks, who made Sex and Violins, are Swedish, and that is the Cotton Eye Joe.SHOW NOTESOutdoor Movie Night Gone Right: Eden's projector plan collapsed due to daylight, so they wheeled the TV outside and screened the 1998 Godzilla with Matthew Broderick — which Eden argues holds up better than its reputation suggests. Friend L contributed an observation about American Godzilla's gender presentation that Peter and Eden both find compelling.Daikon 3 & 4 / Studio Gainax Origin Story: Eden showed friends the legendary fan animations made for the early-'80s Daikon convention circuit — blatant copyright-violating anime crossovers that nonetheless launched the careers of the people who would go on to found Gainax (Neon Genesis Evangelion) and later Trigger (Delicious in Dungeon, Season 2).Roombox 3 Progress: Eden's current miniature diorama project is a laundry room featuring a vending machine, an arcade cabinet, and a complete 1/112-scale Chinese mahjong set (all 144 tiles). Models are being dressed in fabric soft clothing rather than left as bare plastic.LE SSERAFIM New Album: Eden's favorite K-pop act has a new record out. The second track samples La Macarena, which prompted a mild generational crisis at the comic shop when the younger staff noted it predates them.Dungeon Crawler Carl / Discworld Detour: Peter is finishing Book 8 of Dungeon Crawler Carl (narrated by Jeff Hayes, whose per-character voice work Peter genuinely enjoys despite usually disliking that approach) and has resolved to go into Terry Pratchett's Discworld next as a pressure valve from heavy genre fiction.Amon Amarth / Dethklok Concert: Peter drove to Salt Lake for the Amon Amarth/Dethklok tour. Amon Amarth was a highlight — Johan Hegg commanding a full audience Viking rowing session — while Dethklok left Peter cold; the Metalocalypse spectacle on screen keeps the audience at arm's length from the music. Castle Rat opened and was a solid short set.Forza Horizon 6: Peter is ~15 hours in on the Japan-set new installment and finding it an ideal low-commitment diversion. Fits easily into 20-minute sessions or longer stretches.Periphery — A Pale White Dot: New album from the djent-adjacent prog-metal band. Peter's read: fewer peaks but also fewer low points than usual — more consistent, somewhat more middling. Flagged as interesting rather than essential.Bad Album Art Tier List: The main event. Peter used Codex to compile ~45 covers from various internet "worst of" lists into a tier list app, with S = truly worst. Notable rankings: The Faith Tones' Jesus Use Me nearly got its own tier above S; Rednex' Sex and Violins landed S upon discovering the band is Swedish and responsible for the definitive Cotton Eye Joe; Iron Maiden's Dance of Death — described as looking like "Baby's first Blender" — is an A despite being one of their best 21st-century albums; Badfinger's Ass (donkey with headphones, hand holding a carrot) closed the list as a deserved S.Creed Sidebar: Human Clay cover triggers a genuine conversation about Creed's arc — good debut, one-and-a-half good albums, then nothing. Peter credits Alter Bridge as the redemptive outcome.

Jacob and Drew welcome back PaulJPowers.com and Ro... I mean Francisco from the Retro Rewind Podcast plus newcomer Big Salmon to review Flow! Plus the Batman The Animated Series Episodes "You Scratch My Back" and "Never Fear" Retro Rewind Podcast - https://www.firejoystudios.com/s/retro-rewind-podcast PaulJPowersDotCom - https://www.PaulJPowers.com Cel Cast LinkTree - https://linktr.ee/thecelcastpodcast   03:52 - Spoiler-Free Review Begins 11:24 - Spoiler-Filled Review Begins 11:32 - Production Info & Writing Credits 11:58 - Cast & Audio Origins (Recorded from real animals) 14:02 - "Info and Stuff" (Ratings, Streaming, Physical Media) 18:40 - Box Office & Budget Discussion 21:47 - Movie Summary/Plot Walkthrough 25:37 - Trivia (Animal sounds, Blender software, etc.) 28:03 - Final Likes, Dislikes, and Group Discussion 1:13:51 - Culture Box Suggestion: Retro Rewind Podcast 1:14:33 - Patron Shoutouts 1:15:48 - Batman The Animated Series Segment Intro 1:16:04 - Episode: "You Scratch My Back" (Review & Trivia) 1:33:18 - Episode: "Never Fear" (Review & Trivia) 1:47:40 - Social Media & Contact Info  

What happens when the business you planned is not the business you actually build?In this episode, I talk with Brandon McCraney, owner of Old Raleigh Distillery, about leaving corporate life, chasing whiskey, and opening a distillery during COVID.Brandon shares how Old Raleigh started as a blending vision, how a change in North Carolina law turned him into a bar owner, and what it was like opening before he could even sell his own whiskey.We also get into whiskey blending, one-off batches, cask finishes, and why building a bourbon brand takes more than a good product.If you've ever had to pivot because reality wrecked your original plan, this one is worth listening to.Sometimes the business you survive is the business you were supposed to build.

Send us Fan MailWhat are you actually wired to do? Betsy Wills, co-founder of YouScience and author of Your Hidden Genius, joins Daniel Rock and Jason Adair to disrupt the traditional follow your dreams model of career counseling. In this episode, Wills breaks down why self-reported personality tests fall short, how unused biological aptitudes drive adult burnout and how educators can use performance-based data to uncover the unique treasure chest of talent in every single student. Key TakeawaysDefining Aptitudes: Why aptitudes are objective, performance-based, unchanging biological traits rather than learned achievements or temporary interests. The Braided Rope Framework: Balancing personality, exposure-based interests and innate abilities to navigate a changing workforce. The Anatomy of Burnout: Why professional unrest is often caused by an unused aptitude and how avocations can restore career fulfillment. The "Slow Drip" vs. "Blender" Mind: Understanding how traits like idea rate dictate whether you belong in a courtroom, a creative studio or an operating room. The Danger of Mirror Tests: Why self-reported surveys like Myers-Briggs create a boomerang effect that fails to offer rigorous career direction. Democratizing Guidance: Shifting career exploration away from an expensive luxury to a standard tool accessible to all high school students. Resources MentionedYour Hidden Genius: Your Hidden Genius: Leverage Your Natural Talents to Discover Your Career Pathway by Betsy Wills. YouScience Platform: Learn more at YouScience.com. The Southern Regional Education Board is a nonprofit, nonpartisan organization that works with states and schools to improve education at every level, from early childhood through doctoral education and the workforce. Follow Us on Social:FacebookInstagramX

Terrain2STL, Adjusting fit in Blender, 3d Printing and Pet Birds, Stepper Motors for Beginners, 

Link Up w/The Morning Sickness Digitally All Over:Instagram: @hms_98_official, @bosskupd, @bretvesely, @dickToledoX/Twitter: @HMSon98, @DickToledo, @bretveselyFacebook: @HMSKUPDYouTube: @hmspodcast9320, @98kupdRequest/Call in/Wakeup Song line:(IN AZ) 602.585.9800More HMS: holmbergpodcast.com, 98kupd.comEmail: dtoledo@98kupd.com, bvesely@98kupd.com, bbogen@98kupd.comSee Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.

Link Up w/The Morning Sickness Digitally All Over:Instagram: @hms_98_official, @bosskupd, @bretvesely, @dickToledoX/Twitter: @HMSon98, @DickToledo, @bretveselyFacebook: @HMSKUPDYouTube: @hmspodcast9320, @98kupdRequest/Call in/Wakeup Song line:(IN AZ) 602.585.9800More HMS: holmbergpodcast.com, 98kupd.comEmail: dtoledo@98kupd.com, bvesely@98kupd.com, bbogen@98kupd.comSee Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.

LVL Up Expo: Hot off our LVL Up Expo panel, Jim and Them play some catch up on the Kill Switch, the Michael trailer and more!Corey's Michael Premiere Invite & Corey's Publicist: Corey has a long convoluted story regarding his lack of appearance at the Michael movie premiere. Corey's Publicist Sammi goes on a talk show to defend publicists! She's there to advise they are not making bot accounts and attacking people, while she allegedly does that very thing!Corey Doc Deleted Scenes: Thanks to Marcie Hume we have some bonus scenes from the Corey Feldman Vs. The World documentary! A nice peek into the life of the Goblin Ghoul. Also Corey performs for a 108 year old woman at a "biker bar".COREY FELDMAN!, SHOW STOPPER!, LET'S JUST TALK!, DON CHEADLE!, BOOGIE NIGHTS!, JIM AND THEM IS POP CULTURE!, MAYDAY!, OVERTHROW!, STREAMATHON!, DONATE!, 911!, NEW FAMILY!, 22 CHAIN!, SBD!, HDM!, DON DAWN MAY!, DDM!, SHOW TRIP!, KILL SWITCH!, MICHAEL TRAILER!, STARRING COREY FELDMAN!, BEAT IT!, YN!, ORGANIZERS!, MICHAEL PREMIERE!, ALIEN ANT FARM!, 4/20!, MANAGER!, RSVP!, HE LIVED THE STORY!, PEE WEE!, DUSTIN FROM STRANGER THINGS!, KID ACTORS!, FLAWSOME TALK!, JUSTIN BALDONI!, SAMMI ROBIN!, PUBLICIST!, MATT KENNEDY!, MATT SPEAKS TRUTH!, MARCIE HUME!, ANGELS!, HARASS!, PR FIRM!, SAFE!, MIKE COSA!, LEATHERFACE!, TEXAS CHAINSAW!, WRESTLING!, GOSSIPY RAG!, ADRIEN RUMORS!, BREAKUP!, NEW APARTMENT!, FIGHT CLUB!, COCK FRAME!, RULES OF THE BUS!, MARGOT!, BRITTANY!, TOO FAR!, PIZZA EATING!, GROSS!, DARCI!, BLENDER!, GOD!, MESSIAH!, WILD FIRES!, NEVER QUIT!, 108 YEAR OLD LADY!, COVER BAND!, BIKER BAR!, DROPS OF JUPITER!, COLDPLAY!, COREY'S TWITTER!, APARTMENT!, STORIES!,You can find the videos from this episode at our Discord RIGHT HERE!

What This Bride Did With Her Grandmother’s Wedding Gift Turned Out to Be a Huge Mistake!See omnystudio.com/listener for privacy information.

Topics covered include: Kane's parents keeping him off the internet until age eight, pirating way too much software, the origin of the Backrooms, the purgatories we build for ourselves, meeting your heroes after early virality, building 30,000 sq. ft. of sets on a soundstage, Kane accidentally calling his film a "product" and immediately wanting to flee to the woods, shotlisting an entire feature in Blender, renting out a Vancouver theater to watch 2001: A Space Odyssey during prep, James Wan's hopeful advice for young filmmakers, Renate's spiritual curiosity, control variables, and bringing the family together for the theatrical premiere of Backrooms.

Car companies are beginning to use AI tools to radically speed up their development process, which could change the cars we drive forever — and have some big effects on the people who make them now. Verge contributor Tim Stevens explains. Then, The Verge's Hayden Field catches us up on Codex vs. Claude Code, Anthropic vs. the US government, the vibes at OpenAI, and more, before helping answer a question on the Vergecast Hotline (call 866-VERGE11 or email ⁠vergecast@theverge.com⁠!) about whether all the recent tech layoffs are really about AI. Further reading: ⁠The AI-designed car is taking shape | The Verge⁠ ⁠Pentagon strikes classified AI deals with OpenAI, Google, and Nvidia — but not Anthropic⁠ ⁠Google employees ask Sundar Pichai to say no to classified military AI use | The Verge⁠ ⁠Anthropic's new cybersecurity model could get it back in the government's good graces | The Verge⁠ ⁠Microsoft and OpenAI's famed AGI agreement is dead | The Verge⁠ ⁠Here's how the new Microsoft and OpenAI deal breaks down | The Verge⁠ ⁠ChatGPT downloads are slowing — and may cause problems for OpenAI's IPO | The Verge⁠ ⁠Claude can now plug directly into Photoshop, Blender, and Ableton | The Verge⁠ ⁠OpenAI's new security model is for ‘critical cyber defenders' only | The Verge⁠ ⁠Anthropic releases a new Opus model amid Mythos Preview buzz | The Verge⁠ ⁠Jack Dorsey's Block cuts nearly half of its staff in AI gamble | The Verge⁠ Subscribe to The Verge for unlimited access to theverge.com, subscriber-exclusive newsletters, and our ad-free podcast feed.We love hearing from you! Email your questions and thoughts to vergecast@theverge.com or call us at 866-VERGE11. Timestamps are approximate.) 00:00:00 Intro 00:02:00 Today Show Preview 00:04:00 Car Design Primer 00:08:00 AI Speeds Up Design 00:13:00 Clay Models and Craft 00:15:00 Jobs Pipeline Risk 00:18:00 Software Defined Cars 00:20:00 Regulation and Safety 00:27:00 Slate Truck Update 00:34:00 Claude Code vs Codex 00:42:00 OpenAI Vibes Check 00:44:00 PR vs AI Doomerism 00:48:00 Pentagon Deals Exclude Anthropic 00:53:00 Mythos Reality Check 00:56:00 RIP AGI Moment 01:04:00 Hotline AI Layoffs ROI 01:13:00 Wrap Up and Sign Off Learn more about your ad choices. Visit podcastchoices.com/adchoices

Our 243rd episode with a summary and discussion of last week's big AI news!Recorded on 04/29/2026Hosted by Andrey Kurenkov and Jeremie HarrisFeel free to email us your questions and feedback at andreyvkurenkov@gmail.com and/or hello@gladstone.aiRead out our text newsletter and comment on the podcast at https://lastweekin.ai/In this episode:OpenAI released GPT-5.5 with strong coding-oriented improvements, a system card discussing chain-of-thought monitorability and misalignment testing, higher pricing than GPT-5.4, and notable quirks like a system-prompt warning about “goblins.”xAI launched Grok Voice Think Fast 1.0, claiming large benchmark leads for real-time voice agents and reporting major Starlink customer-support automation and sales conversion impact.DeepSeek open-sourced DeepSeek V4 (Pro and Flash) featuring MoE scaling and 1M-token context via hybrid/compressed attention changes, while Tencent released Hunyuan 3 preview with weaker benchmark performance; a new long-horizon agent benchmark (Clawmark) shows low task success rates.Major business, legal, and policy updates include Google's planned up-to-$40B investment and 5GW compute commitment to Anthropic, Meta's AWS Gravitron deal and China blocking Meta's Manus acquisition, a revamped OpenAI–Microsoft agreement, ongoing Musk–OpenAI trial developments, and new safety/security research on sabotage, document degradation under delegation, and bit-flip attacks.Timestamps:(00:00:10) Intro / Banter(00:02:00) News Preview(00:02:26) Response to listener comments(00:02:55) SponsorsTools & Apps(00:05:55) OpenAI Unveils Its New, More Powerful GPT-5.5 Model - The New York Times(00:23:33) xAI Launches grok-voice-think-fast-1.0: Topping τ-voice Bench at 67.3%, Outperforming Gemini, GPT Realtime, and More - MarkTechPost(00:29:00) Claude can now plug directly into Photoshop, Blender, and Ableton | The VergeProjects & Open Source(00:29:38) China's DeepSeek releases preview of long-awaited V4 model as AI race intensifies(00:47:05) Tencent Unveils Hy3 preview; Model Enhances Agent Capabilities and Real-World Usability - Tencent 腾讯(00:50:14) ClawMark: A Living-World Benchmark for Multi-Turn, Multi-Day, Multimodal Coworker AgentsApplications & Business(00:53:03) Google Plans to Invest Up to $40 Billion in Anthropic(00:56:26) Meta will use hundreds of thousands of AWS Graviton chips(00:59:51) China blocks Meta's $2 billion takeover of AI startup Manus(01:01:45) OpenAI shakes up partnership with Microsoft, capping revenue share payments(01:07:13) Elon Musk Testifies of AI Risk at Trial, Says OpenAI Tried to ‘Steal' a Charity - WSJ(01:11:50) Judge rejects DOJ bid to delay Anthropic appeal in Pentagon dispute(01:14:42) Google's Gemini can now run on a single air-gapped server — and vanish when you pull the plug(01:19:07) DeepMind's David Silver just raised $1.1B to build an AI that learns without human data | TechCrunchPolicy & Safety(01:22:47) Evaluating whether AI models would sabotage AI safety research(01:28:59) LLMs Corrupt Your Documents When You Delegate(01:32:50) Temporal Sparse Autoencoders: Leveraging the Sequential Nature of Language for Interpretability(01:39:53) Memorandum on Adversarial Distillation of American AI Models(01:41:41) Teen boys are dating their AI chatbots—and experts warn it could kill their careers | Fortune(01:43:57) Announcing the Anthropic Economic Index Survey(01:45:21) Scoop: CISA lacks access to Anthropic's MythosSynthetic Media & Art(01:48:03) Taylor Swift Files to Trademark Voice and Likeness to Protect Against AI MisuseResearch & Advancements(01:49:15) Maximal Brain Damage Without Data or Optimization: Disrupting Neural Networks via Sign-Bit FlipsSee Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.

The Wall Street Journal reported OpenAI missed its end-of-year billion-active-user target. Is the AI bubble actually popping or is the panic overblown?  This week on AI For Humans, the AI bubble panic hit a fever pitch after a Wall Street Journal report revealed OpenAI missed its weekly user, monthly revenue, and end-of-year billion-active-user targets. CFO Sarah Friar reportedly told peers she's worried OpenAI won't be able to pay for future compute contracts if revenue doesn't accelerate, and the board is now scrutinizing Sam Altman's deals more closely.  AI stocks crashed, with Oracle, AMD, and CoreWeave all sinking on the news. Anthropic is eating into OpenAI's market share on coding and enterprise. We dig into whether the bubble is actually popping or whether this is a panic that conflates AI capability with the business of AI. Spoiler: we don't think it's over.  Plus, DeepSeek v4 launched and made surprisingly small waves. OpenAI's deal with Microsoft expanded to other clouds. Meanwhile, AI itself is absolutely not slowing down: Tom Cruise is running faster than ever in a viral video, OpenAI dropped a new Chappie-style voice interaction, Claude got a Blender connector for 3D modeling, NVIDIA released Nemotron 3 Nano Omni with 30B parameters and 256K context, and Talkie is an LLM trained entirely on 1880s text.  OPENAI MISSED ITS NUMBERS. IS THIS THE BUBBLE? YES. NO. SHRUG EMOJI? #ai #ainews #openai  Come to our Discord: https://discord.gg/muD2TYgC8f Join our Patreon: https://www.patreon.com/AIForHumansShow AI For Humans Newsletter: https://aiforhumans.beehiiv.com/ Follow us for more on X @AIForHumansShow Join our TikTok @aiforhumansshow To book us for speaking, please visit our website: https://www.aiforhumans.show/   // Show Links // WSJ: OpenAI Misses Key Revenue and User Targets https://www.wsj.com/tech/ai/openai-misses-key-revenue-user-targets-in-high-stakes-sprint-toward-ipo-94a95273 AI Stocks Sink on OpenAI News (Yahoo Finance) https://finance.yahoo.com/markets/article/oracle-amd-and-coreweave-stocks-sink-after-report-says-openai-missed-sales-user-targets-130600628.html DeepSeek v4 Launch https://x.com/deepseek_ai/status/2047516922263285776?s=20 Sam Altman: OpenAI's Microsoft Deal Expands to Other Clouds https://x.com/sama/status/2048755148361707946?s=20 Kwindla's Smart Take on AI's Importance https://x.com/kwindla/status/2049161481149935668?s=20 Tom Cruise Runs Faster: Viral AI Video https://x.com/Le_Chuck_81/status/2049027447304196297?s=20 New Chappie-Style Voice Interaction From OpenAI https://x.com/OpenAIDevs/status/2048871260512473385?s=20 Blender Connector From Claude https://x.com/claudeai/status/2049143438281445811?s=20 NVIDIA's Nemotron 3 Nano Omni Announcement https://x.com/NVIDIAAI/status/2049159441870717428?s=20 Talkie: LLM Trained on 1880s Text https://x.com/status_effects/status/2048878495539843211?s=20  

This week on The Whiskey Trip podcast, Big Chief sits down with Preston Wall for a ride that goes beyond the glass, diving into service, sacrifice, and the power of real connection. Frontline Heroes Outdoors (FHO) is a Texas-born 501(c)(3) nonprofit dedicated to supporting active duty military, veterans, law enforcement, and first responders—along with their families, with a strong focus on spouses of deployed members and Gold Star Families. Through outdoor adventures and shared experiences, FHO creates a space where camaraderie is rebuilt, purpose is rediscovered, and no one has to carry the weight alone.  On the first half, Big Chief pours a 12-year Old Charter Oak French Oak while Preston shares how it all began. Aged in rare French Oak, this bourbon delivers rich sweetness, subtle oak influence, and a silky, refined texture that sets the tone for a powerful story of mission and impact.  On the second half, Big Chief cracks into the new Blender's Edition 01 from Knob Creek, the first in a limited series focused on the art of blending. This release brings bold notes of dark vanilla, honey, and rich spice. At 10 years old, 106 proof, and priced at $39.99, it raises the question of whether Jim Beam is signaling a shift—bringing real value back to the shelf.  Preston also sips on one of FHO's barrel picks from Southern Spirits Collective, highlighting a key part of their mission. These hand-selected barrels aren't just great whiskey—they're tied directly to giving back. Each pick helps support the heroes and families FHO serves, turning every pour into something bigger than the glass. It's whiskey with purpose, building community one bottle at a time.  This episode is about more than great whiskey. It's about honoring those who serve, building something meaningful, and supporting a mission that truly matters. Take the Ride with Big Chief. Cheers

Welcome back to today's Friday Review where I'll be breaking down the best of the week!     I'll be sharing specifics on these topics:     Stainless Steel Blender Jar (product review) Recovery Determines Results (tip of the week) Early Morning Workers (research) Viagra Use for Leigh Syndrome      For all the details tune in to today's Cabral Concept 3724 – Enjoy the show and let me know what you thought!   - - - For Everything Mentioned In Today's Show: StephenCabral.com/3724 - - - Get a FREE Copy of Dr. Cabral's Book: The Rain Barrel Effect - - - Join the Community & Get Your Questions Answered: CabralSupportGroup.com - - - Dr. Cabral's Most Popular At-Home Lab Tests: > Complete Minerals & Metals Test (Test for mineral imbalances & heavy metal toxicity) - - - > Complete Candida, Metabolic & Vitamins Test (Test for 75 biomarkers including yeast & bacterial gut overgrowth, as well as vitamin levels) - - - > Complete Stress, Mood & Metabolism Test (Discover your complete thyroid, adrenal, hormone, vitamin D & insulin levels) - - - > Complete Food Sensitivity Test (Find out your hidden food sensitivities) - - - > Complete Omega-3 & Inflammation Test (Discover your levels of inflammation related to your omega-6 to omega-3 levels) - - - Get Your Question Answered On An Upcoming HouseCall: StephenCabral.com/askcabral - - - Would You Take 30 Seconds To Rate & Review The Cabral Concept? The best way to help me spread our mission of true natural health is to pass on the good word, and I read and appreciate every review!  

It's This Week in Bourbon for April 17th 2026. Sazerac has reportedly approached Brown-Forman regarding a potential merger, MGP stops distilling operations at Limestone Branch Distillery and Lux Row Distillers, and Oscar Mayer has announced its first bacon innovation in five years with the launch of Maple Bourbon Bacon with Evan Williams. Show Notes: Sazerac has approached Brown-Forman with a rival merger bid, complicating Brown-Forman's ongoing "merger of equals" talks with Pernod Ricard Sazerac has rebranded its La Vergne, Tennessee, site as the AJ Bond Distillery and will launch its first Tennessee whiskey this summer MGP Ingredients is idling distilling operations at two Kentucky facilities for at least 12 months starting May 1, 2026, to address market oversupply Kentucky Artisan Distillery Program launched a new single barrel program offering smaller 15- and 25-gallon barrels to increase accessibility for consumers A new 60-seat speakeasy focused on bourbon history and Prohibition-era legends will open beneath Louisville's Hotel Distil on May 5, 2026 Give 270's 19th Whiskey Wednesdays round, the "Bourbon Buddy Bonus," features weekly raffles for rare whiskey through June 24, 2026 GALLO has acquired Four Roses Bourbon from Kirin Holdings, bringing the historic brand back under U.S. family ownership The Kentucky Bourbon Trail hosted 2.7 million visitors in 2025, with 80% traveling from outside Kentucky and 62% reporting household incomes over $100,000 Foley Family Wines & Spirits launched Gambit No. 6, a 6-year-old Kentucky Straight Bourbon finished in six different barrel types, priced at $69.99 James B. Beam Distilling Co. launched the Blender's Edition 01, a 10-year-old, 106-proof release priced at $44.99 New Riff/Rhinegeist released DUET American Whiskey, a 6-year-old, 111.2-proof blend of malted barley, raw barley, and rye priced at $79.99 Hard Truth French Oak Finished Bourbon, a sweet mash whiskey finished in French oak casks, is available now for $69.99 Rabbit Hole Raceking's new 6-year-old, 5-grain mash bill bourbon debuts on April 14, 2026, at its Louisville distillery Buffalo Trace is releasing the Single Oak Rye Bourbon ($74.99) and the Experimental Collection Low Entry Proof Wheated Bourbon ($46.99) this April Oscar Mayer partnered with Evan Williams to launch a new maple bourbon-cured bacon as part of an annual innovation strategy Koopers is releasing 225 bottles of a 7-year-old, Cognac-cask finished rye on April 11, 2026, priced at $80 per bottle The third "Greats of the Gate" bottle, honoring Hall of Fame horse Northern Dancer, releases April 16, 2026, to support Kentucky nonprofits Learn more about your ad choices. Visit megaphone.fm/adchoices