Podcasts about Chips

"This is not just an important day for Nvidia, but an important day for the tech sector," says James Demmert. With Nvidia (NVDA) reporting earnings after Wednesday's closing bell, he believes CEO Jensen Huang needs to discuss a clear roadmap for the company that includes guidance and addressing a constrained supply chain. While China remains a key question mark, James calls the current stock price a "great opportunity" for buyers. Tom White turns to an example options trade for the Mag 7 leader. ======== Schwab Network ========Empowering every investor and trader, every market day.Options involve risks and are not suitable for all investors. Before trading, read the Options Disclosure Document. http://bit.ly/2v9tH6DSubscribe to the Market Minute newsletter - https://schwabnetwork.com/subscribeDownload the iOS app - https://apps.apple.com/us/app/schwab-network/id1460719185Download the Amazon Fire Tv App - https://www.amazon.com/TD-Ameritrade-Network/dp/B07KRD76C7Watch on Sling - https://watch.sling.com/1/asset/191928615bd8d47686f94682aefaa007/watchWatch on Vizio - https://www.vizio.com/en/watchfreeplus-exploreWatch on DistroTV - https://www.distro.tv/live/schwab-network/Follow us on X – https://twitter.com/schwabnetworkFollow us on Facebook – https://www.facebook.com/schwabnetworkFollow us on LinkedIn - https://www.linkedin.com/company/schwab-network/About Schwab Network - https://schwabnetwork.com/about

R. “Ray” Wang and Ryan Shrout preview what to watch in Nvidia (NVDA) earnings. Ray says this report could be a “reset” for the AI trade on both sides. Ryan is watching for comments around future products and next-gen chip specs. He's also looking for customer demand metrics for its Reuben chips. Ray comments on what China revenue could look like and Jensen Huang's plans for the future. ======== Schwab Network ========Empowering every investor and trader, every market day.Options involve risks and are not suitable for all investors. Before trading, read the Options Disclosure Document. http://bit.ly/2v9tH6DSubscribe to the Market Minute newsletter - https://schwabnetwork.com/subscribeDownload the iOS app - https://apps.apple.com/us/app/schwab-network/id1460719185Download the Amazon Fire Tv App - https://www.amazon.com/TD-Ameritrade-Network/dp/B07KRD76C7Watch on Sling - https://watch.sling.com/1/asset/191928615bd8d47686f94682aefaa007/watchWatch on Vizio - https://www.vizio.com/en/watchfreeplus-exploreWatch on DistroTV - https://www.distro.tv/live/schwab-network/Follow us on X – https://twitter.com/schwabnetworkFollow us on Facebook – https://www.facebook.com/schwabnetworkFollow us on LinkedIn - https://www.linkedin.com/company/schwab-network/About Schwab Network - https://schwabnetwork.com/about

What happens when a high-performance race shop decides to jump into aerospace and defense manufacturing — and goes all in? In this episode of Machine Shop Mastery, I sit down with Marty Moran of TKO Precision Machining and TKO Motorsports in Reno, Nevada. What started in 2008 as a motorsports-focused shop building custom race cars and high-end components evolved into a serious aerospace and defense manufacturing operation about eight years ago. Marty shares how the team leveraged deep motorsports and aerospace experience to enter defense manufacturing, earn AS9100 certification, and build a thriving 15-machine shop. But what stands out most isn't just their growth — it's their culture. Communication is constant. Training is intentional. Hiring is rigorous. And everyone is expected to succeed. We talk about workforce development, cross-training machinists into race crew roles, the realities of AS9100 compliance, building depth through mentorship, and the painful ERP lesson that ultimately led them to ProShop. Marty also shares why aerospace certification doesn't just open doors — it makes you a better shop. If you're trying to build a resilient, team-driven shop in today's manufacturing environment, this conversation is packed with insight. You will want to hear this episode if you are interested in... (0:00) Introducing Marty Moran and the origins of TKO Motorsports (2:18) How the business evolved from a race shop into contract machining (4:02) Launching TKO Precision Machining as a focused aerospace operation (6:05) Current shop size, equipment mix, and aerospace capabilities (9:40) How the motorsports division operates alongside contract machining (14:35) Integrating machinists into race team operations (19:10) Breaking into aerospace and defense manufacturing (22:40) Starting with prototype work to build long-term customers (25:05) Navigating ITAR and NIST 800-171 compliance (29:20) Revenue diversification between motorsports and defense work (32:05) Building culture through cross-training and accountability (36:10) Hiring philosophy and what TKO looks for in new employees (41:20) Peer-driven hiring process and extended evaluation periods (45:00) Developing operators into machinists through internal training (48:55) Measuring spindle utilization and operational efficiency (52:05) Communication rhythm and leadership accessibility (54:30) Lessons learned from ERP implementation and systems discipline (56:20) Advice for smaller shops on training, retention, and culture (58:00) Final reflections on teamwork and what's next for TKO Resources & People Mentioned Grow your top and bottom-line with CliftonLarsonAllen Why we love SMW Autoblok for workholding Mark your calendars and come see us at IMTS 2026 Connect with Marty Moran Connect on LinkedIn TKO Precision Machining Connect With Machine Shop Mastery The website LinkedIn YouTube Instagram Subscribe to Machine Shop Mastery on Apple, Spotify Audio Production and Show Notes by - PODCAST FAST TRACK

Ireland's National Competence Centre in Semiconductors (I-C3), a significant milestone in Ireland's commitment to semiconductor innovation and European collaboration under the European Chips Act, invites startups and SMEs to lead the future of chips innovation. I-C3 will focus on startups and SMEs by providing access to essential resources, including funding pathways, training, design tools and pilot line facilities. Its mission is to empower Ireland's startups and SMEs in the semiconductor sector with hands-on access to design, production, funding and training to accelerate innovation and growth in Ireland's semiconductor sector. National Competence Centre in Semiconductors for Startups Commenting on the launch, Peter Burke TD, Minister for Enterprise, Tourism and Employment said: "As a hub for the semiconductor ecosystem, my Department is delighted that I-C3 will ensure that opportunities as part of the Chips for Europe Initiative are accessible for businesses of all sizes within the industry, along with bringing greater diversity of expertise and depth of innovation to the knowledge base of the semiconductor ecosystem in Europe. I-C3's launch is another significant milestone in the delivery of Silicon Island: Ireland's National Semiconductor Strategy. "With this launch, my Department is very excited about I-C3's ability to empower Irish SMEs to scale internationally, drive innovation across the semiconductor ecosystem and create high-value jobs. I-C3 will also facilitate the development of skills and talent, and build on our strengths by enhancing the relationship between infrastructure, industry, and RD&I capability to ensure Ireland leads in advanced manufacturing and chip design." Co-ordinated by Tyndall National Institute and supported by the Department of Enterprise, Tourism and Employment (DETE) through Enterprise Ireland, with co-funding secured from the European Union under the Chips Joint Undertaking (Chips JU), I-C3 is a consortium comprising Tyndall National Institute, a research flagship of University College Cork (UCC), MCCI, MIDAS Ireland, NovaUCD, and University College Dublin. The new I-C3 Competence Centre is one of 30 being established across 27 EU countries to strengthen Europe's semiconductor ecosystem. The initiative builds on Ireland's vibrant and extensive semiconductor industry comprising over 130 indigenous and foreign subsidiary companies, employing over 20,000 people, part of a 175,000-person strong broader ICT sector with overall exports of €13.5 billion worth of products annually. Multinational leaders such as Intel, Apple, Qualcomm, AMD, and Analog Devices have long invested in Irish R&D. I-C3 aims to further elevate Ireland's global standing in semiconductor innovation. Professor William Scanlon, CEO, Tyndall, said: "I?C3 plays a key role in delivering Ireland's Semiconductor Strategy, Silicon Island, and it is fantastic to see the centre operational and actively supporting Irish start?ups and SMEs to accelerate and scale their businesses. I?C3 is helping companies across all sectors that use semiconductor technologies to secure investment, access specialist training, and connect to European pilot lines." Joe Healy, Divisional Manager, Research, Innovation and Infrastructure at Enterprise Ireland said: "With the support of I-C3, Ireland is set to double the number of people employed in semi-conductor startups and SMEs by 2030. The centre will act as a catalyst for innovation, collaboration, and growth, ensuring that Irish stakeholders, from academia to industry, can fully participate in the Chips for Europe Initiative." About Tyndall National Institute Tyndall is a leading European deep-tech research centre in integrated ICT (Information and Communications Technology) materials, devices, circuits and systems and a research flagship of University College Cork. Tyndall is Ireland's largest Research and Technology Organisation (RTO) specialising in both electronics and photonics. Tyndall works...

De wereld staat onder hoogspanning. Trump is opnieuw aan de macht in Amerika en de oorlogen in Gaza en Oekraïne houden de wereld in hun greep. Wat merken we daarvan in Nederland? Investico deed het afgelopen jaar meerdere onderzoeken naar een plek waar dat misschien wel als eerste voelbaar is: de wetenschap. Samenwerking is voor universiteiten een goudmijn: ze komen verder door het delen van kennis en inzichten met de beste wetenschappers over de hele wereld. Maar wanneer kan je dat niet meer verantwoorden? Hoe gaan de Nederlandse universiteiten om met een steeds ondemocratischere wereld? Kan je bijvoorbeeld nog samenwerken met collega's die gecensureerd worden? En wat doe je als je mede-onderzoeker meevecht in de oorlog in Gaza? In deze speciale aflevering van Speurwerk vertellen Investico-journalisten Bijou van der Borst, Sofyan El Bouchtili en Emiel Woutersen samen met hoofdredacteur Thomas Muntz over hoe de effecten van de veranderende geopolitiek direct voelbaar zijn in de Nederlandse wetenschap. Ze bespreken drie verschillende onderzoeken waarin word gekeken naar hoe Nederlandse universiteiten zich verhouden tot een steeds illiberalere wereld. Ze deden deze onderzoeken samen met Machteld Veen, Emma van Bergeijk en Martijn Roessing voor Trouw, De Groene Amsterdammer, Nu.nl en het hoger onderwijs persbureau. Presentatie & montage: Sylvana van den Braak & Michelle Salomons Muziek & eindmix: Pepijn Buitenhuis

Plus: Apple to move some Mac Mini desktop computer production from Asia to Houston. And Anthropic announces updates to Claude Cowork. Julie Chang hosts. Learn more about your ad choices. Visit megaphone.fm/adchoices

Live Feb 24, 2026 | Yaron Brook ShowAI Impact; Russia War; Iran–Will He?; US Chips; Prediction Markets; Welfare Fraud | Yaron Brook ShowThe Yaron Brook Show is Sponsored by:-- The Ayn Rand Institute (https://www.aynrand.org/starthere)-- Energy Talking Points, featuring AlexAI, by Alex Epstein (https://alexepstein.substack.com/)-- Express VPN (https://www.expressvpn.com/yaron)-- Hendershott Wealth Management (https://www.youtube.com/watch?v=X4lfC...) https://hendershottwealth.com/ybs/-- Michael Williams & The Defenders of Capitalism Project (https://www.DefendersOfCapitalism.com)Join this channel to get access to perks: / @yaronbrook Like what you hear? Like, share, and subscribe to stay updated on new videos and help promote the Yaron Brook Show: https://bit.ly/3ztPxTxSupport the Show and become a sponsor: / yaronbrookshow or https://yaronbrookshow.com/ or / yaronbrookshow Or make a one-time donation: https://bit.ly/2RZOyJJContinue the discussion by following Yaron on Twitter (https://bit.ly/3iMGl6z) and Facebook (https://bit.ly/3vvWDDC )Want to learn more about Ayn Rand and Objectivism? Visit the Ayn Rand Institute: https://bit.ly/35qoEC3#IranProtests #RussiaUkraineWar #Tariffs #Individualism #Capitalism #Geopolitics #China #WesternCivilization #objectivismBecome a supporter of this podcast: https://www.spreaker.com/podcast/yaron-brook-show--3276901/support.

Plus: Novo Nordisk is set to cut prices of GLP-1's by as much as half next year. And consumer confidence was up for February. Anthony Bansie hosts. Sign up for WSJ's free What's News newsletter. An artificial-intelligence tool assisted in the making of this episode by creating summaries that were based on Wall Street Journal reporting and reviewed and adapted by an editor. Learn more about your ad choices. Visit megaphone.fm/adchoices

This week, Brendon, Federico, and John tackle the RAM shortage and its effect on the handheld industry before digging into Cocoon 2.0, an excellent Android front-end for your emulators. Also available on YouTube here. Links and Show Notes AYN Odin 3 Ultra Available Again for Pre-Order AYN Odin 3 Ultra RAM Shortage & Industry Impact The out-of-stock Steam Deck is the latest victim of the RAM & storage shortage Cocoon 2.0 Frontend Cocoon 2.0 Subscribe to NPC XL NPC XL is a weekly members-only version of NPC with extra content, available exclusively through our new Patreon for $5/month. Each week on NPC XL, Federico, Brendon, and John record a special segment or deep dive about a particular topic that is released alongside the "regular" NPC episodes. You can subscribe here: https://www.patreon.com/c/NextPortableConsole Leave Feedback for John, Federico, and Brendon NPC Feedback Form Credits Show Art: Brendon Bigley Music: Will LaPorte Follow Us Online On the Web MacStories.net Wavelengths.online Follow us on Mastodon NPC Federico John Brendon Follow us on Bluesky NPC MacStories Federico Viticci John Voorhees Brendon Bigley Affiliate Linking Policy

Schickt uns euer Feedback zur EpisodeWas passiert, wenn ein Netzwerk zur Lebensader wird? Das Internet war mal eine utopische Idee – ein Versprechen von grenzenlosem Wissen und demokratischer Teilhabe. Mittlerweile ist es selbstverständliche Infrastruktur. Und leider eben auch ein hart umkämpftes Schlachtfeld.In dieser Folge haben sich Mary-Jane und Patrick die Tech-Journalistin und Co-Host des ARD-Podcasts „Der KI Podcast", Marie Kilg, eingeladen – und zusammen schauen wir uns an, was da gerade schiefläuft, aber auch, wo echte Chancen liegen.Wir ziehen die Linie von Brockhaus zu Wikipedia und von dort zu algorithmischen Feeds, die Aufmerksamkeit belohnen und Wut verstärken, während Konsens leise verschwindet. Wir reden über Medien zwischen Rendite und Aufklärung, über Selbstzensur, Machtdrift und polarisierende Anreize, die Debatten verformen, bis niemand mehr fragt, was stimmt – sondern nur noch, was klickt.Dann drehen wir die Perspektive: Kann dieselbe KI, die das anheizt, uns auch helfen? Offene Modelle, Datensouveränität, transparente Trainingsdaten – das klingt nach Nerd-Hobby, ist aber die Voraussetzung dafür, dass Unternehmen, Verwaltungen und Zivilgesellschaft Kontrolle und Werte zurückgewinnen. Wir schauen auch dahin, wo KI ganz praktisch einen Unterschied machen kann – etwa in Verwaltungen, die mit wachsender Komplexität und knappen Ressourcen umgehen müssen. Gleichzeitig reden wir Klartext über Grenzen: Energie, Chips, reale Kosten und die Frage, ob agentische Systeme echte Wertschöpfung liefern – oder nur neue Abhängigkeiten.Die Zukunft ist nicht determiniert. Zwischen Plattformmacht und Gemeinwohl entscheidet sich gerade, was wir heute bauen, regulieren und fördern. Es liegt an uns.

The following article of the Tech industry is: “Chips Made in Mexico: The Beginning of Technological Sovereignty” by Alejandro Franco Rodríguez, CEO, QSM Semiconductores (AA2414)

Record Numbers Shatter Post-Pandemic Expectations Six years after the pandemic first disrupted American life, a troubling trend emerges across rural Virginia. The Blue Ridge Area Food Bank now serves approximately 180,000 people every month—a staggering 39,000 more than the pandemic's peak. Les Sinclair, the organization's Communications and PR Manager, reveals this sobering reality during a recent conversation on The Valley Today with host Janet Michael. Initially, food bank officials believed the pandemic would represent the worst crisis they'd ever face. When government assistance programs temporarily lifted many families out of poverty, demand dropped slightly to around 141,000 monthly visits. However, this optimism proved short-lived. "We thought the numbers would never go up beyond the pandemic max," Les explains. "That just didn't pan out." Instead, inflation took hold with devastating consequences. While prices soared across every sector, wages failed to keep pace. Consequently, more working families find themselves unable to afford basic necessities, forcing them to seek food assistance for the first time in their lives. A Massive Rural Footprint The Blue Ridge Area Food Bank operates across an impressive territory that spans 25 counties and eight cities throughout Virginia. Stretching from Winchester and Frederick County in the north to beyond Lynchburg and Bedford County in the south, the organization covers approximately 12,000 square miles—roughly the size of Maryland or one-third of Virginia's total area. To manage this vast region effectively, the food bank maintains four strategic warehouse locations. Their headquarters sits in Verona, just outside Staunton, while additional distribution centers operate in Winchester, Charlottesville, and Lynchburg. Notably, the Winchester facility alone serves Frederick, Clarke, Fauquier, Warren, Shenandoah, Page, and Rappahannock Counties, including the densely populated Loudoun County. Moreover, the organization represents a groundbreaking experiment in food banking. When founded in 1981, most food banks concentrated on urban areas where dense populations made distribution easier. The Blue Ridge Area Food Bank, however, pioneered rural food distribution—a critical distinction since nine out of ten food-insecure Americans live in rural communities rather than urban centers. The Partnership Model That Makes It Work The food bank functions as a sophisticated logistics operation, partnering with Feeding America nationally and hundreds of local food pantries regionally. Les compares their role to a Walmart warehouse, buying food by the truckload and storing massive quantities. Meanwhile, local pantries like Winchester CCAP serve as the "customer-facing" locations, directly distributing food to families in need. This partnership proves essential for reaching scattered rural populations. "We couldn't do what we do without them," Les emphasizes. "They couldn't do what they do without us." Furthermore, the organization sources food from diverse channels. Retail grocers contribute 36% of donations through partner pickup programs, where pantries collect excess inventory directly from stores like Food Lion, Kroger, and Giant. Additionally, the USDA provides government-purchased food from American farmers, while large manufacturers donate products with misprinted labels or excess inventory. Local and regional farmers also contribute fresh produce to the network. The Grocery Store Challenge Recently, however, the retail partnership faced unexpected pressure. During October and November, and again during winter snowstorms, consumers cleared grocery store shelves completely. When stores have no excess inventory, they have nothing left to donate. Compounding this challenge, grocery chains have become remarkably efficient at predicting demand. Using AI technology, they now anticipate that shoppers will buy strawberry Pop-Tarts before storms and adjust inventory accordingly. While this efficiency benefits retailers and consumers, it reduces the surplus available for food banks. Simultaneously, USDA food supplies have dropped 30% year-over-year, forcing the food bank to purchase more food directly. Although they cannot fully replace the high-quality proteins and vegetables the government typically provides, they continue prioritizing nutritious options for their partner pantries. Shattering Misconceptions About Food Pantry Users Perhaps the most persistent myth surrounding food insecurity involves who actually needs assistance. Many people assume food pantry visitors are simply lazy and should "get a job." The reality, however, tells a dramatically different story. Most people seeking food assistance are working. They're trying to improve their lives but living on financial margins so thin that a single unexpected expense creates crisis. In fact, more than a quarter of the food bank's guests visit only once per year—they simply need help getting over a temporary hump. Les shares the story of a convenience store worker who injured her wrist on the job. Unable to work while waiting for workers' compensation, she has zero income and cares for a paralyzed son. She's not lazy—she's injured, uninsured temporarily, and desperately trying to survive until she can return to work. Even when workers' compensation arrives, it typically covers only 70% of regular wages and takes considerable time to process. For families living paycheck to paycheck, missing even one payment creates cascading financial disasters. The Government Shutdown Ripple Effect Currently, partial government shutdowns compound these challenges. Federal workers, particularly TSA agents, continue reporting to work without paychecks. They still pay for childcare, gas, and other necessities, but many receive payment only monthly—making it extraordinarily difficult to stretch resources from one paycheck to the next. Contrary to popular belief, landlords cannot always wait patiently for delayed rent payments. Many landlords depend on rental income to pay their own mortgages. When a tenant misses a $2,000 rent payment, the landlord must still cover their mortgage. Moreover, the economic impact extends far beyond government employees. When federal workers stop dining out, restaurants lose business. Wait staff lose tips. Restaurant owners order less food from suppliers like Sysco. Truck drivers haul fewer loads. The entire economic system suffers. Sarah Cohen of Route 11 Chips experienced this firsthand. During COVID and government shutdowns, her sales to DC cafes plummeted because federal workers weren't coming to the office for lunch. These ripple effects reach deep into Virginia's economy, affecting businesses and workers far from the capital. The Impossible Choice: Heat or Eat Winter brings particularly cruel dilemmas for struggling families. Les recently spoke with William, a roofer injured on the job who lives in a mobile home with his dog, Cocoa. Unable to afford heating, William and Cocoa "just sort of curl up" together while he waits for surgeries that will allow him to return to work. Another woman caring for three disabled grandchildren faces $400 monthly electric bills. With both she and her husband experiencing serious health issues and the children's parents out of the picture, they constantly struggle with the impossible choice between heating their home and feeding their family. These aren't isolated cases. Across the food bank's service area, families regularly face this devastating decision. When $600 heating bills arrive after cold snaps, many choose to keep the lights on and visit food pantries to feed their families. Food as Medicine: A Holistic Approach The Blue Ridge Area Food Bank takes a progressive stance on nutrition, viewing food as medicine rather than mere sustenance. They prioritize fresh produce, which comprised 30% of their distribution last year, because they understand that proper nutrition helps people thrive. Nutritious food keeps medical bills down across entire communities. Children pay better attention in school when properly nourished. People can manage chronic illnesses and diseases through better nutrition. Conversely, when families can only afford high-calorie processed foods, they face increased health risks despite consuming adequate calories—debunking the myth that overweight individuals cannot be food insecure. Additionally, access to food reduces stress, which itself functions as a health intervention. When people live on the edge of a financial cliff, they cannot make good long-term decisions. They're too focused on simply not falling. However, when food security removes one major stressor, families can step back from that precipice and begin making better choices for their futures. Quality Food for Everyone Another common misconception suggests that food bank offerings are somehow subpar. In reality, the food distributed through this network maintains high-quality standards. While well-meaning donors sometimes contribute items like ramen noodles during food drives, the bulk of distributed food comes from retail grocers, USDA programs, and direct purchases of nutritious items. The food bank specifically prioritizes produce because people crave fresh fruits and vegetables. Although produce represents one of the most expensive food categories—often making it a luxury for families on tight budgets—the organization believes everyone deserves access to healthy, nutritious food regardless of their economic circumstances. How Communities Can Help Fortunately, community members have multiple ways to support this critical mission. Volunteering provides valuable assistance, and notably, many food bank guests themselves volunteer, giving back to the community that supported them during difficult times. Financial donations prove particularly effective. Just $1 helps provide more than three meals, meaning $10 supplies a month of meals for someone in need, while $100 provides 300 meals. The food bank's purchasing power and logistics expertise amplify every dollar donated. Beyond time and money, advocacy matters tremendously. Currently, the Federation of Virginia Food Banks—representing all seven food banks across the state—works to promote "food as medicine" initiatives with the state legislature. Community members can support these efforts through the food bank's website at BRAFB.org/actnow or BRAFB.org/getinvolved. Finally, social media engagement amplifies the message. Following the food bank's social media accounts, resharing posts, and commenting helps spread awareness that hunger relief remains an urgent community need. Finding Help When You Need It For individuals and families currently struggling with food insecurity, Les offers an important message: "You're not alone, and we are here with you. We are here to walk with you through this challenge in your life." The food bank's website features an easy-to-use food finder tool. Visitors to BRAFB.org can click "Find Food," enter their address, and immediately see all nearby pantries with contact information, open hours, and everything needed to access food quickly. Alternatively, Virginians can call 211 for phone-based assistance connecting them with local resources. A Community Responsibility As this conversation reveals, food insecurity affects far more people than most realize—one in nine people across the food bank's service area. These aren't strangers or statistics; they're neighbors, coworkers, and community members facing temporary crises that could happen to anyone. The Blue Ridge Area Food Bank stands ready to help, but they cannot do it alone. Through partnerships with local pantries, support from community donors and volunteers, and advocacy for systemic solutions, the organization continues fighting to ensure everyone has enough to eat. In Janet Michael's words, it's "a responsibility I do not take lightly"—and neither should any of us.

Medusa and Cam look to tie the threads between a long awaited sequel and it's take on spirituality... but instead just added a fun lil guy! Look at that lil guy! Our Flipped Script: Let's make it a co-op! Intermission: 1:07: -1:19: Suggested Topics: Psyop Your Parents, Shakled to the First Player, Fart Sniffy, AGDQ, Playin' Jazz, CreeEeePy Crawlers 

Die allererste Fastnacht in Mainz endet für Basti im viel zu engen Mario-Kostüm mit einer Küche voller Kollateralschäden. Währenddessen zweifelt anredo nach einem beunruhigenden Zahnarzt-Besuch an seiner eigenen Ersatzteilfähigkeit. BastiMasti hätte sich seinen ersten Rosenmontag zur Fastnacht in Mainz definitiv anders vorgestellt. Nicht nur sein Freundeskreis macht schlapp, auch das neue Heimatdomizil Mainz weiß nicht so recht zu begeistern. So endet der nachgemachte Kölner Karneval in der nachgemachten Domstadt nicht nur mit Erbrochenem, sondern mit einem Polyester-Overall, der wirklich jede Ader sichtbar macht. Jede. Einzelne. Die Zuchthengst-Ader inklusive. Der Versuch, das Mario-Kostüm gemächtsfähig zu machen, scheitert kläglich. Stattdessen wird kurzfristig auf Cowboy umgerüstet, während draußen Menschen in Hauseingänge urinieren und ein Typ mit dem Teppichreiniger von Rossmann durch die Stadt irrt. Zwischen Weinschorle-Frühstück, Dixi-Klo-Eskapaden und einem Notfall-Küchen-GAU mit LocknLock-Frischhaltedosen, Nicer Dicer und einer Pfanne voller Elend wird klar: Alkohol ist vielleicht doch kein Lifestyle für die Mainz-Era von Sebastian Mast. Parallel kämpft Ex-Internetstar anredo mit seinem lauten Kiefer. Ist er ein abnormales Knack-und-Back-Brötchen? Braucht er eine Knirsch-Schiene oder muss der komplette Kiefer ausgetauscht werden wie ein defektes IKEA-Teil? Und warum schmeckt sein Notvorrat-Wasser aus den neuen Soda Stream Flaschen plötzlich besser als je zuvor? Dazu gibt es eine Exkursion in die Welt von Xavier Naidoo und die große Frage, ob man im Ernstfall lieber Lay’s Chips oder Karnevalskostüme bunkern sollte… Diese und alle anderen Episoden #rundfunk17 findet ihr unter anderem bei Apple Podcasts, Spotify, Deezer und als RSS-Feed.

James Chai, Visiting Fellow at ISEAS and former policy advisor to Malaysia's Ministry of Economy, joins Jeremy Au to unpack how Malaysia is repositioning itself in an era defined by AI, semiconductors, and geopolitical rivalry. They explore the country's shift from oil, gas, and plantations toward advanced manufacturing, examine how decades of semiconductor clustering built a quiet but durable export engine, and discuss why Malaysia is now doubling down on data centers and rare earths. The conversation covers US China competition over chip supply chains, the strategic importance of fabrication and GPU ecosystems, and how rare earth processing may represent the most underappreciated leverage point in the global tech stack. James also explains why execution, not ambition, will determine whether Malaysia can capture long term value from these emerging industries. 02:30 Malaysia balances growth with redistribution: The strategy is to raise high value industries like semiconductors and rare earths while lifting the bottom 40 percent through social protection. 05:42 Semiconductor strength came from decades of compounding: Intel and other multinationals anchored early manufacturing, and local engineers accumulated expertise that later spun into globally competitive firms. 10:18 Clusters beat subsidies alone: Tight networks of engineers, spin offs, and long term continuity allowed Malaysia's chip ecosystem to survive volatility and keep upgrading. 21:05 China uses constraint as strategy: By limiting access to high end Nvidia GPUs, Beijing forces domestic firms to innovate faster and close critical design gaps. 29:45 Chips are not oil: Frontier GPUs power model training, but most real world AI use relies on inference, meaning older chips retain value longer than markets assume. 37:22 Data centers create investment headlines but unclear spillovers: Billions flow into Malaysia, yet long term value depends on whether local firms capture supply chain and technology capabilities. 44:10 Rare earth processing is the real choke point: Deposits are global, but China controls the complex multi step processing chain, making chemistry and technology control more strategic than mining alone. Watch, listen or read the full insight at https://www.bravesea.com/blog/james-chai-rare-earth-power Get transcripts, startup resources & community discussions at www.bravesea.com WhatsApp: https://whatsapp.com/channel/0029VakR55X6BIElUEvkN02e TikTok: https://www.tiktok.com/@jeremyau Instagram: https://www.instagram.com/jeremyauz Twitter: https://twitter.com/jeremyau LinkedIn: https://www.linkedin.com/company/bravesea English: Spotify | YouTube | Apple Podcasts Bahasa Indonesia: Spotify | YouTube | Apple Podcasts Chinese: Spotify | YouTube | Apple Podcasts Vietnamese: Spotify | YouTube | Apple Podcasts #MalaysiaEconomy #Semiconductors #RareEarths #DataCenters #USChinaTech #Geopolitics #AIStrategy #SupplyChains #IndustrialPolicy #BRAVEpodcast

No episódio desta semana, Pedro tem companhia de Kiki Rivotti no seu estilo Steve Jobs descalço. Os dois debatem grandes temas da sociedade atual - partilhar localização com namorados e proibir o Instagram a menores de 16 anos - mas também outros assuntos, como o traço de personalidade "chegar sempre atrasado", a aparição de Toy no Rio de Janeiro e a introdução do novo serviço Uber iQos.(00:00) Intro(00:23) Pessoas que chegam sempre atrasadas: porquê?(05:34) PTM descobre novo tipo de tosse(06:30) Um grande elogio recebido no ginásio(08:34) Como reagir a comida estragada em restaurante?(12:27) Toy aparece no Rio de Janeiro(15:04) Beber chá com francesinha faz sentido?(17:06) Estar no Porto sabe a viagem em cidade europeia(19:02) Desumificador funciona mesmo?(21:12) Justin Bieber estreia boxers novos todos os dias(22:44) Campanha "Relationchip" foi eficaz?(24:28) Partilhar localização com amigos(28:47) Faz sentido saber passwords de outras pessoas?(32:38) Quiz sobre privacidade em relações(34:59) ChatGPT não sabe melhorar textos(38:36) CNN partilha tiktok de AI(40:50) Aranhas gigantes na ponte no dia 1 de Abril de 2013(43:14) TVI lança vídeo de aniversário em AI(45:59) Limitação de redes sociais a menores de 16 anos(51:56) Kiki quer implementar Uber iQos(55:21) Posição dos partidos políticos sobre a limitação de redes sociais a menores de 16 anos(56:30) Crianças de 10 anos no TikTok(1:01:20) Ficar sem redes sociais durante 1 ano

Free League is bringing back the iconic world of Trudvang as a standalone tabletop roleplaying game, *Dragonbane: Trudvang*, which is fully compatible with their current, award-winning edition of *Dragonbane*. The setting, rooted in Nordic mythology, features shadowed woods, dreadful beasts, and endless mystery. The core rulebook will adapt and expand the *Dragonbane* ruleset for the *Trudvang* setting, including new kin, professions, heroic abilities, and unique magic for vitner weavers and dimwalkers. The Kickstarter campaign offers four hardback volumes in Swedish and English, with early access Beta PDFs a few months after the campaign concludes. Modiphius and Humble Bundle have released a massive Fallout Tabletop Mega Bundle, which includes content for both the *Fallout Tabletop RPG* and the *Fallout: Wasteland Warfare* miniatures skirmish game. The bundle offers over $1k worth of items for $25, with a significant draw being the STL files for miniatures, allowing 3D printing enthusiasts to create various terrain pieces, iconic Bobbleheads, cars, and other items from the Wasteland. Even without a 3D printer, the bundle provides a robust selection of PDF rulebooks, including the RPG Core Rulebook, Settler’s and Wanderer’s Guides, NPC packs, the *Winter of Atom* campaign, and *Wasteland Warfare* expansions like *Liberty Prime*. A portion of the proceeds from the bundle, which has already raised over $14k, supports the American Civil Liberties Union (ACLU). MyMiniFactory announced its acquisition of Thingiverse, the world’s oldest and largest 3D printing community, adding it to its family alongside YouMagine. MyMiniFactory’s decade-long focus has been on building a sustainable platform for creators, having paid out over $100 million to designers. The acquisition aims to support the underserved Thingiverse community by providing creators with the necessary tools, support, and business models. As part of MyMiniFactory’s 2025 “SoulCrafted initiative” to protect human creativity from AI-generated content, Thingiverse will now adopt a zero-tolerance stance on AI-generated designs. The platforms will remain independent, and the open sharing culture, including existing free models, will be preserved. Mantic Games has announced they will be publishing a new *Ghost in the Shell* Tabletop Roleplaying Game in 2026. This new RPG is based solely on the original manga by Shirow Masamune. It features a bespoke design by Alessio Cavatore and Zak Barouh, combining fast-paced, narrative-driven mechanics with the philosophical depth of the source material. The game is set in a near-future world of cybernetics and political intrigue, where players design their own trainee agents in Section 9 and features rules for combat and hacking, a rich armoury of gear, and the first mission, *Lost Patriot*. This is the second *Ghost in the Shell* TTRPG due in the same year, following the crowd-funded *Ghost in the Shell Arise TTRPG* by Mana Project Studio, which is based on the anime. #trudvang #fallout #myminifactory #mantic #gits Dragonbane Trudvang: https://www.kickstarter.com/projects/1192053011/dragonbane-trudvang-the-legend-returns Fallout MEGA Bundle: https://humblebundleinc.sjv.io/L0aArV Warmachine on MyMiniFactory: https://mmf.io/upturned Mantic Companion App: https://companion.manticgames.com/ Use our Referral code: MCTXEE Support us by Shopping at Miniature Market (afilliate link): https://miniature-market.sjv.io/K0yj7n Support Us by Shopping on DTRPG (afilliate link): https://www.drivethrurpg.com?affiliate_id=2081746 Matt’s DriveThruRPG Publications: https://www.drivethrurpg.com/browse.php?author=Matthew%20Robinson https://substack.com/@matthewrobinson3 Chris on social media: https://hyvemynd.itch.io/​​ Jeremy's Links: http://www.abusecartoons.com/​​ http://www.rcharvey.com ​​Support Us on Patreon: https://www.patreon.com/upturnedtable Give us a tip on our livestream: https://streamlabs.com/upturnedtabletop/tip​ Donate or give us a tip on Paypal: https://www.paypal.com/ncp/payment/2754JZFW2QZU4 Intro song is “Chips” by KokoroNoMe https://kokoronome.bandcamp.com/

The playoffs are coming — and if you're not planning your schedule now, you're already behind.Nate and Blake break down Weeks 21–24 in detail, highlighting the teams with the best fantasy playoff schedules, the trade targets to acquire before it's too late, the streamers who can win you categories, and the playoff strategies that separate contenders from champions.This is your fantasy hockey playoff cheat code. Let's get those CHIPS!

Daniel is joined by Linley Gwennap, technology analyst and author of the new book “How AI Really Works: The Models, Chips, and Companies Powering a Revolution.” Linley was the long-time editor of Microprocessor Report and chaired the popular Linley Processor Conferences. Dan explores what impact AI is having on the market and… Read More

Xsight Labs talks their latest SDN X2 network switch and E1 DPU chips with the GreyBeards

Ken and Lima look back on their younger days while Anthony gives tips on how to stay fit into your 40s

Shoot us a Text.Episode #1273: GM is staying lean to outmaneuver the next sales slowdown. AI's appetite for memory chips could spark a new supply squeeze across autos and tech. Retailers are proving that telling better stories sells.Show Notes with links:General Motors is rewriting its inventory playbook, running 30–40% leaner and hoping that tighter supply, stronger cash flow, and faster decision-making could turn the next cycle into a competitive advantage.S&P Global Mobility forecasts U.S. sales down 2.5% to 15.8M units as affordability and softer EV demand weigh on the market.GM is targeting a 50–60 day supply versus the pre-pandemic 100+ days.Leaner inventory gives GM more flexibility to adjust incentives in a downturn without crushing profitability.Dealers have felt the squeeze, especially on affordable models, prompting GM to stage select Trax and Trailblazer units at ports to speed delivery.CFO Paul Jacobson summed up the strategy: “It's easier to do when you have less inventory in the system because you can just respond much more quickly.”Just when the auto industry thought it survived the chip crisis, here comes round two—this time powered by AI. Data centers are devouring global memory supply, forcing automakers to brace for tighter supply, higher costs, and potential production headaches.AI data centers are soaking up global DRAM and memory production, with Western Digital and Seagate already sold out of most 2026 capacity.Memory chip prices have jumped 90% quarter-over-quarter, prompting PC makers like Dell to raise prices 15–20%.Tesla's Elon Musk says the solution may be vertical integration: “We're going to hit a chip wall if we don't do the fab.”Retailers are doubling down on something we at More Than Cars know well—storytelling sells. Brands are shifting from simply stocking products to crafting narratives that spark emotion, build loyalty, and turn casual shoppers into long-term fans.Nordstrom says department stores no longer “introduce” brands—they help tell their story and build deeper consumer connection.Five Below credits curated social storytelling—merchandising and marketing working together—for stronger engagement with younger shoppers.Under Armour's Kevin Plank says brands must inspire emotion: “The world does not need another capable apparel and footwear manufacturer. The world needs hope and they need a dream.”Today's show is brought to you by ESi-Q. ESi-Q measures employee satisfaction and provides actionable insight into what's driving employee engagement and turnover - before employees leave.Join Paul J Daly and Kyle Mountsier every morning for the Automotive State of the Union podcast as they connect the dots across car dealerships, retail trends, emerging tech like AI, and cultural shifts—bringing clarity, speed, and people-first insight to automotive leaders navigating a rapidly changing industry.Get the Daily Push Back email at https://www.asotu.com/ JOIN the conversation on LinkedIn at: https://www.linkedin.com/company/asotu/

Chips from a Calcutta Workshop: Comparative Religion in Nineteenth Century India (Cambridge University Press, 2025) explores the development and nature of comparative religion in nineteenth-century India. It focuses on the ideas and intellectual currents behind a range of thinkers who explored comparative religion in India, drawing on a variety of inspirations from Indian religions. Rather than emanate out of a European Christian set of politics as in the Western world, comparative religion emerged out of religious reform movements, including the Brāhmo Samaj in Bengal and the Arya Samaj in the Punjab. With chapters on Rammohan Roy, Debendranath Tagore, Keshab Chandra Sen, and Swami Vivekananda, the book includes a re-evaluation of familiar figures alongside lesser-known thinkers within an intellectual history of modern Indian comparative religion. Neilesh Bose is Professor of History at the University of Victoria. Learn more about your ad choices. Visit megaphone.fm/adchoices Support our show by becoming a premium member! https://newbooksnetwork.supportingcast.fm/new-books-network

GW27 Preview is here. Quick turnaround coming off an eventful DGW, @FPLUSABrian & @FPLUSABux recap Arsenal & Wolves playing twice, debate Gabriel TC perspectives and highlight the massive fixture swings on the horizon.The banter is top, the league is at max chaos & we hope this episode makes your FPL arrows green with joy!

In episode 105, we finally get the stream dialed and dive straight into hands‑on Bitcoin mining and open-source hardware updates. We share the latest on Ember One: a sneaky IO voltage domain bug uncovered by Mujina dev Ryan led to a desk‑side hardware fix that's now pushing ~2 TH/s (target is 3.6 TH/s across 12 chips with proper cooling). We unpack chip and hashboard design lore—from stacked voltage domains and reliability in long chains to the insider politics at big silicon shops like Intel. We talk why selling chips openly matters, how spec sheets unlock real builder momentum, and why third‑party system builders (think Epic Blockchain) can grease the skids between chipmakers and end products.We cover Mujina's trajectory toward a universal, Linux‑first, open firmware for miners—auto‑detect dreams vs config realities—and near‑term support for Ember One's Intel boards and existing Antminers. We riff on home‑miner UX, remote monitoring, and agent/LLM tooling (cron‑job‑with‑superpowers, heartbeats, MCP integrations) to tune, alert, and manage miners. There's buzz around FutureBit's Apollo 3 (likely Auradine chips), open vs lawyered licenses, and the path from FPGA teaching rigs to community‑designed ASICs. We celebrate community hashing on the 256F HydroPool hash‑dash, solo‑block wins, and Heat Punk Summit prep (immersion hot tub included). Plus, a call to action: support developer freedom at change.org/billandkeonne. It's a dense, builder‑first session on chips, firmware, agents, and bringing practical hashrate‑heat products to life.

The City Bar's Presidential Task Force on AI and Digital Technologies hosts today's podcast on President Trump's:  Winning the Race, America's AI Action Plan. Task Force co-chair Jerome Walker is joined by task force members Matthew Bacal (Davis Polk), Azish Filabi (American College of Financial Services), Robert Mahari (Stanford Codex), and Evan Abrams (Steptoe), to review the plan's three pillars and key action steps. Pillar One (“Accelerate AI Innovation”) is described as largely deregulatory, including agency review of rules and certain FTC/FCC actions, with targeted concerns such as ideological bias and synthetic media in the legal system, plus investments in open-source/open-weight models, data, interpretability, evaluations, and government/DoD adoption. Pillar Two (“Build American AI Infrastructure”) focuses on the physical side of AI—permitting for data centers and fabs, energy and grid expansion, semiconductors, water for cooling, workforce training, cybersecurity, and “security by design,” while anticipating trade-offs and litigation. Pillar Three (“Lead in International AI Diplomacy and Security”) balances support for exporting US “full stack” AI with tighter national security controls, including stronger export-control enforcement and participation in international bodies primarily to counter China. The conversation closes with suggestions for improving the plan by strengthening trust, safety/rights considerations, and maintaining flexibility as AI capabilities evolve. If you are interested in learning more about emerging AI developments and policy, join us for the 2026 Artificial Intelligence Conference on June 18 to hear from industry experts and connect with leading legal professionals across the field. 00:00 Trump's 2025 AI Action Plan: Big Goals, Short Document, 3 Pillars 03:23 Pillar One Preview: 15 Action Steps to ‘Accelerate AI Innovation' 09:16 Meet the Panel + Setting Up the Pillar One Deep Dive 11:21 Pillar One Explained: Deregulation, Free Speech, Data Sharing, Evaluations, and Trust 18:33 Key Takeaways for Stakeholders: Business, Finance, Civil Society, and Tech 23:57 Which Pillar One Steps Matter Most? Sequencing, Competitiveness, and Data Access 27:52 Pillar Two: The Physical Side of AI—Energy, Chips, Data Centers 36:32 Critical Infrastructure Security: Physical Risks, Cyber Threats & ‘Security by Design' 37:14 Data Poisoning Explained: How Training Data Can Be Manipulated at Scale 38:00 Workforce Training at Scale: From Trades to Semiconductor Talent Pipelines 38:52 Wrapping Pillar Two: China Competition, Speeding Projects, and Ranking Priorities 40:34 What Lawyers & Judges Need to Know About Pillar Two (Red Tape, Legal Tech, Litigation) 45:30 Pillar Three Overview: Balancing Global AI Leadership with National Security Controls 50:05 Pillar Three Priorities by Industry: Export Controls, Frontier Evaluations & Data Center Risk 58:56 Why Engage International AI Bodies? Countering China and Filling the Leadership Vacuum 01:03:20 Trump vs. Biden Narratives: Competition vs. Safety—What Should Change in the Plan? 01:07:38 Panel Advice to Improve the Action Plan: Rights Framework, Nimble Policy, Safety & Research Funding

Chips from a Calcutta Workshop: Comparative Religion in Nineteenth Century India (Cambridge University Press, 2025) explores the development and nature of comparative religion in nineteenth-century India. It focuses on the ideas and intellectual currents behind a range of thinkers who explored comparative religion in India, drawing on a variety of inspirations from Indian religions. Rather than emanate out of a European Christian set of politics as in the Western world, comparative religion emerged out of religious reform movements, including the Brāhmo Samaj in Bengal and the Arya Samaj in the Punjab. With chapters on Rammohan Roy, Debendranath Tagore, Keshab Chandra Sen, and Swami Vivekananda, the book includes a re-evaluation of familiar figures alongside lesser-known thinkers within an intellectual history of modern Indian comparative religion. Neilesh Bose is Professor of History at the University of Victoria. Learn more about your ad choices. Visit megaphone.fm/adchoices Support our show by becoming a premium member! https://newbooksnetwork.supportingcast.fm/intellectual-history

Chips from a Calcutta Workshop: Comparative Religion in Nineteenth Century India (Cambridge University Press, 2025) explores the development and nature of comparative religion in nineteenth-century India. It focuses on the ideas and intellectual currents behind a range of thinkers who explored comparative religion in India, drawing on a variety of inspirations from Indian religions. Rather than emanate out of a European Christian set of politics as in the Western world, comparative religion emerged out of religious reform movements, including the Brāhmo Samaj in Bengal and the Arya Samaj in the Punjab. With chapters on Rammohan Roy, Debendranath Tagore, Keshab Chandra Sen, and Swami Vivekananda, the book includes a re-evaluation of familiar figures alongside lesser-known thinkers within an intellectual history of modern Indian comparative religion. Neilesh Bose is Professor of History at the University of Victoria. Learn more about your ad choices. Visit megaphone.fm/adchoices Support our show by becoming a premium member! https://newbooksnetwork.supportingcast.fm/south-asian-studies

Chips from a Calcutta Workshop: Comparative Religion in Nineteenth Century India (Cambridge University Press, 2025) explores the development and nature of comparative religion in nineteenth-century India. It focuses on the ideas and intellectual currents behind a range of thinkers who explored comparative religion in India, drawing on a variety of inspirations from Indian religions. Rather than emanate out of a European Christian set of politics as in the Western world, comparative religion emerged out of religious reform movements, including the Brāhmo Samaj in Bengal and the Arya Samaj in the Punjab. With chapters on Rammohan Roy, Debendranath Tagore, Keshab Chandra Sen, and Swami Vivekananda, the book includes a re-evaluation of familiar figures alongside lesser-known thinkers within an intellectual history of modern Indian comparative religion. Neilesh Bose is Professor of History at the University of Victoria. Learn more about your ad choices. Visit megaphone.fm/adchoices Support our show by becoming a premium member! https://newbooksnetwork.supportingcast.fm/indian-religions

Chips from a Calcutta Workshop: Comparative Religion in Nineteenth Century India (Cambridge University Press, 2025) explores the development and nature of comparative religion in nineteenth-century India. It focuses on the ideas and intellectual currents behind a range of thinkers who explored comparative religion in India, drawing on a variety of inspirations from Indian religions. Rather than emanate out of a European Christian set of politics as in the Western world, comparative religion emerged out of religious reform movements, including the Brāhmo Samaj in Bengal and the Arya Samaj in the Punjab. With chapters on Rammohan Roy, Debendranath Tagore, Keshab Chandra Sen, and Swami Vivekananda, the book includes a re-evaluation of familiar figures alongside lesser-known thinkers within an intellectual history of modern Indian comparative religion. Neilesh Bose is Professor of History at the University of Victoria. Learn more about your ad choices. Visit megaphone.fm/adchoices Support our show by becoming a premium member! https://newbooksnetwork.supportingcast.fm/religion

There are no RBT's on the air... Chips crunch like apples to addict you! We miss sleepovers, let's bring them back David Genat reveals survivor secrets Dear Abby, my situationship is KILLING me We get the inside scoop from locals... Is Ed coming to Ipswich today?! See omnystudio.com/listener for privacy information.

In the News The First Android 17 Beta is now Available on Pixel Devices T-Mobile Adds Free New Service Hate Saying ‘Hey Google' to your Nest devices? Key Points from Recent Updates to OpenAI's Privacy Policy  NASA's Latest Attempt to Launch Artemis 2 Didn't Go As Planned SpaceX Crew-12 Mission Latest News NASA Puts 21-Year-Old Spacecraft on Pause   ITPro Series with Benjamin Rockwell Why Your Company Is Suddenly Talking About Data Retention Policies From the Tech Corner Nvidia's Momentum is Running into Real-World Limits Mini Retro Compact Cameras Technology Chatter with Benjamin Rockwell and Marty Winston AI and Chips, and the Quantity of Chips in a Small Space  

Market update for Wednesday February 18, 2026 Check out the Public app for incredible investing tools and to support the show (LINK)Follow us on Instagram (@TheRundownDaily) for bonus content and instant reactions.In today's episode, Zaid covers:Berkshire Hathaway slashes Amazon stake, trims AppleGold and silver pull back after recent runMeta strikes multiyear deal to buy millions of Nvidia AI chipsUber invests $100M in autonomous vehicle charging hubsCaesars jumps on Vegas reboundSandisk falls after Western Digital announces $3.1B share saleRecord CEO turnover hits Corporate America, and leaders are getting younger

En este episodio de VG Daily, Valentina Orduz y Andre Dos Santos revisan una sesión bursátil con índices casi planos, pero con tono claramente defensivo, tecnología y software bajo presión, utilities y real estate al alza, y tasas largas cerca de 4%.Luego analizan el reporte de Palo Alto Networks, donde el crecimiento de ingresos y utilidades contrasta con una guía de utilidades más débil por los costos de integrar varias adquisiciones en ciberseguridad e IA, en un contexto en el que el mercado castiga cualquier señal de presión en márgenes, mientras Meta firma un acuerdo multianual para comprar millones de chips de Nvidia y rediseñar sus data centers alrededor de su plataforma de cómputo de IA.Finalmente, conectan este entorno con la economía real a partir del recorte de guía de General Mills, que habla de la debilidad del consumidor y ve caídas en volúmenes en categorías básicas, y lo enmarcan dentro del patrón histórico del segundo año de un primer mandato republicano, típicamente más flojo para el S&P 500 en la primera parte del año.

Following the weekend' FA Cup football and last night's 5th round draw Suj and James revaluate the outlook for Chip usage for the rest of the season, and though current projections remain unchanged there's lots that could change in the coming weeks and that may force many FPL managers to move away from the current template strategy. What will happen to the Manchester City v Crystal Palace fixture? Could Gameweek 33 be the only remaining Double Gameweek? Should Bench Boost be used now or held for later? It's all covered, including alternative 'out' chip strategies if at short notice Gameweek 32 would look bad for a Wildcard followed by a Gameweek 33 Bench Boost, why certain fixtrures could be forced into certain weeks, if non European teams could have postponements go into random weeks and more, including why the next big event that could alter the landscape is the scheded April TV selections... Tomorrow on Planet FPL: Clash of the Correspondents, Aston Villa v Leeds United with Lee Jackson & Ed Salinger Today on Patreon: UCL Fantasy Preview (IT+) & Euro Focus (AT) The full Planet FPL schedule for this week can be found via this post: https://www.patreon.com/posts/150897034 Want to become a member of our FPL community and support the Podcast?  Join us on Patreon: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.patreon.com/planetfpl⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ Follow James on Twitter/x: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://twitter.com/PlanetFPLPod⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ Follow Suj on Twitter/x: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://twitter.com/sujanshah⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ Follow Clayton on Twitter/x: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://twitter.com/claytsAFC⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ Follow David on Twitter/x: https://x.com/PlanetFPLHunter Follow Nico on Twitter/x: https://twitter.com/nico_semedo Subscribe to our YouTube channel: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.youtube.com/@PlanetFPL⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ Like us on Facebook: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.facebook.com/planetfpl⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ Follow us on Instagram: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.instagram.com/planetfpl⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ #FPL #FantasyPremierLeague #ChipStrategy Learn more about your ad choices. Visit podcastchoices.com/adchoices

Hey Pickles!Here's what's coming up in today's show!In this week's Y Files, Is Russia putting brain chips into pigeons?Here's the article: https://peakd.com/hive-196387/@davideownzall/brain-chipped-pigeons-in-russia-exploiting-sentient-beings-as-flying-machinesIn our Noteworthy segment, recent good news out of Amsterdam concerning meat advertising.Here's the article: https://veganhorizon.substack.com/p/amsterdam-bans-meat-advertisingAnd, in Our Main Topic, we're taking a look at the new dietary guidelines coming out of the Trump administration.Here's the article: https://www.pcrm.org/news/news-releases/new-dietary-guidelines-were-written-authors-strong-ties-food-industry-doctorsWe have a new Listener Shout Out, and lots of our usual shenanigans!Thanks so much for spending some of your day with us.Much love, Sam & ChristineSend us a text! We can't respond, but we'd love to hear from you!Support the showJoin Our Patreon https://www.patreon.com/CompassionandcucumbersSign Up For Our Newsletterhttps://www.compassionandcucumbers.comOur YouTube https://www.youtube.com/@compassioncucumbersveganpod/videos72 Reasons To Be Vegan *paid link https://amzn.to/3W8ZwsUVisit Our Website https://www.compassionandcucumbers.comSam's Etsy https://www.etsy.com/shop/CucumberCraftworks

Tragic Tipp City Shooting; Best Fish & Chips in Dayton and Fish Fry Fridays; Robert Duvall dies at 95; Curling Controversy; Nancy Guthrie Mystery; Smart Underwear; GPS fails.

We are back with more news and stories you will only find here! Plus comedy! Roll a Meldrick and enjoy the moment!

In this issue of Moneycontrol Editor's Picks, our focus: the India AI Summit where venture capitalist Vinod Khosla predicted AI will make jobs obsolete, Union Minister Ashwini Vaishnaw said India will start commercial production of memory chips soon, Union MoS Pemmasani Chandra Sekhar assured telecom backbone is ready to take on AI workload and more. Also read about tax scrutiny, trade negotiations and property prices as we bring you all the top headlines from the day. Tune in!

Last week Tottenham finally parted ways with Thomas Frank with the club in 16th place and looking nervously over their shoulder at Nottingham Forest and West Ham. Former Juventus manager Igor Tudor has been given the managerial job until the end of the season. Here James is joined by fellow Tottenham fan Ricky Saunders for a discussion on the managerial change, why blame should also be aimed at those running the football club and concerns around a chronic injury list. Plus, what we can expect from Tudor tactically and stylistically and if the job is just waiting for Mauricio Pochettino's return... but would he even take it if the club went down? Tomorrow on Planet FPL: s9 ep40 Chips & Strategy Continued Today on Patreon: The Patreon QNA (BT+) & The Fan View Extra Time with Ricky Saunders (AT) The full Planet FPL schedule for this week can be found via this post: https://www.patreon.com/posts/150897034 Want to become a member of our FPL community and support the Podcast?  Join us on Patreon: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.patreon.com/planetfpl⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ Follow James on Twitter/x: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://twitter.com/PlanetFPLPod⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ Follow Suj on Twitter/x: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://twitter.com/sujanshah⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ Follow Clayton on Twitter/x: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://twitter.com/claytsAFC⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ Follow David on Twitter/x: https://x.com/PlanetFPLHunter Follow Nico on Twitter/x: https://twitter.com/nico_semedo Subscribe to our YouTube channel: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.youtube.com/@PlanetFPL⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ Like us on Facebook: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.facebook.com/planetfpl⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ Follow us on Instagram: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.instagram.com/planetfpl⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ #THFC #Tottenham #Spurs Learn more about your ad choices. Visit podcastchoices.com/adchoices

- "Ride the Wave, Build the Future: Scientific Computing in an AI World", by Dongarra, Reed, Gannon - Call for National Moonshot Program for future HPC systems - DOE Genesis Mission, 26 Challenges for National Science and Technology - NSF $100M National Quantum and Nanotechnology Infrastructure, NQNI - State of The Quantum Computing Industry - Los Alamos National Laboratory Center for Quantum Computing [audio mp3="https://orionx.net/wp-content/uploads/2026/02/HPCNB_20260216.mp3"][/audio] The post HPC News Bytes – 20260216 appeared first on OrionX.net.

Plus: Goldman Sachs' top lawyer Kathryn Ruemmler steps down amid the Epstein files fallout. And Coinbase posts a big loss as Bitcoin's fall drags down the wider crypto market. Daniel Bach hosts. Sign up for WSJ's free What's News newsletter. Learn more about your ad choices. Visit megaphone.fm/adchoices

VanMan: ⁠https://vanman.shop/⁠Book a call: ⁠https://remnantfinance.com/calendar⁠ ! Out Print the Fed with 1% per week: https://remnantfinance.com/optionsEmail us at info@remnantfinance.com or visit https://remnantfinance.com for more informationFOLLOW REMNANT FINANCEYoutube: @RemnantFinance (https://www.youtube.com/@RemnantFinance )Facebook: @remnantfinance (https://www.facebook.com/profile.php?id=61560694316588 )Twitter: @remnantfinance (https://x.com/remnantfinance )TikTok: @RemnantFinanceDon't forget to hit LIKE and SUBSCRIBEIf you've been in the health-conscious space online, you've seen Van Man products everywhere — tallow balm, eggshell tooth powder, fluoride-free mouthwash. But most people don't know the story behind the brand.In this episode, Jeremy Ogorek sits down with Hans to talk about losing everything in a New York tech startup, moving back in with his mom, buying a van, and accidentally stumbling into a health brand that's now replacing every product in your bathroom — and soon, your pantry too. We also get into the "everything is a lie" awakening, why fluoride was his first red flag, what's actually in the products you put on your skin, and how he's now selling $6 grass-fed smash burgers out of a restaurant in Pacific Beach that keeps selling out.If you've been rethinking what you put on and in your body, this one's for you.Chapters: 00:00 – Opening segment 01:25 – Van's background: CPA, quitting his first job, joining a NYC tech startup 05:15 – The startup collapse: $8M raised, celebrity investors, and losing everything 08:55 – Fluoride as the first red flag and the origin of the eggshell tooth powder 14:05 – How the tallow balm was born and why it went viral 19:00 – "Your skin is a mouth" — the philosophy behind Van Man products 21:25 – Product lineup: deodorant, sunscreen, bug balm, soap, shampoo, eye cream 30:30 – The Van Man restaurant in Pacific Beach: $6 grass-fed burgers 36:00 – The business model: restaurants, gas stations, and movie theaters as product "stunts" 43:25 – Other clean brands: Masa Chips, Orum, Rosie's Chips 53:00 – Vaccines, home birth, and the broader health awakening 57:00 – What's next: tallow popcorn, clean Snickers bars, cough drops, and an RFK collab1:04:15 – Closing segmentKey Takeaways:Tallow isn't a trend — it's a return to what worked for thousands of years. People are reporting cleared rosacea, vanishing acne, and healed scars from a balm made of five ingredients you could eat. Meanwhile, the dermatologist-recommended steroid creams weren't solving the same problems in a decade.Your skin is your largest organ, and it absorbs what you put on it. If you wouldn't eat the ingredients in your lotion or deodorant, ask yourself why you're comfortable rubbing them into your skin — especially in high-absorption areas like your armpits.Fluoride was the first domino. It's the only non-opt-in medication — it's in your tap water, your toothpaste, and it's free. Once you ask why they care so much about your cavities, the rest of the questioning begins.The restaurant isn't really about the restaurant. Van Man Burgers in Pacific Beach sells $6 grass-fed smash burgers at near break-even. The real play is getting clean products in front of new customers. Every "stunt" — restaurant, gas station, movie theater — is a storefront for the mission.You don't need permission to start. Van went from credit card debt and a van to building a brand, a restaurant, and a product line — all by following his gut, tweeting his thoughts, and making products he wanted to use himself. The XP comes from doing, not reading.

On this episode of In Pursuit with Rich Froning, we’ve got the guys from High Ground Beef Chips in the studio, Dylan Larson and Josh McCandless. It started with a problem that kicked it all off: finding real, lightweight nutrition on the move between deployments, long days, and no patience for filler ingredients. Dylan breaks down why they built High Ground as a meat chip: simple label, big protein, whole-food fuel that actually works in the backcountry. From there, the conversation goes into why hunting/training with the boys can be a reset when life gets loud. We also get the meaning behind the name High Ground and their bigger goal: supporting Gold Star families and telling the stories that shouldn’t get forgotten. Connect with Rich Froning MeatEater on Instagram, Facebook, Twitter, Youtube, and Youtube Clips Subscribe to The MeatEater Podcast Network on YouTubeSee omnystudio.com/listener for privacy information.

AT&T's $2000 shell, ZFS Scrubs and Data Integrity, FFS Backups, FreeBSD Home Nas, and more. NOTES This episode of BSDNow is brought to you by Tarsnap and the BSDNow Patreon Headlines One too many words on AT&T's $2,000 Korn shell and other Usenet topics Understanding ZFS Scrubs and Data Integrity News Roundup FFS Backup FreeBSD: Home NAS, part 1 – configuring ZFS mirror (RAID1) 8 more parts! Beastie Bits The BSD Proposal UNIX Magic Poster Haiku OS Pulls In Updated Drivers From FreeBSD 15 FreeBSD 15.0 VNET Jails Call for NetBSD testing Tarsnap This weeks episode of BSDNow was sponsored by our friends at Tarsnap, the only secure online backup you can trust your data to. Even paranoids need backups. Feedback/Questions Gary - Links Send questions, comments, show ideas/topics, or stories you want mentioned on the show to feedback@bsdnow.tv Join us and other BSD Fans in our BSD Now Telegram channel

From rewriting Google's search stack in the early 2000s to reviving sparse trillion-parameter models and co-designing TPUs with frontier ML research, Jeff Dean has quietly shaped nearly every layer of the modern AI stack. As Chief AI Scientist at Google and a driving force behind Gemini, Jeff has lived through multiple scaling revolutions from CPUs and sharded indices to multimodal models that reason across text, video, and code.Jeff joins us to unpack what it really means to “own the Pareto frontier,” why distillation is the engine behind every Flash model breakthrough, how energy (in picojoules) not FLOPs is becoming the true bottleneck, what it was like leading the charge to unify all of Google's AI teams, and why the next leap won't come from bigger context windows alone, but from systems that give the illusion of attending to trillions of tokens.We discuss:* Jeff's early neural net thesis in 1990: parallel training before it was cool, why he believed scaling would win decades early, and the “bigger model, more data, better results” mantra that held for 15 years* The evolution of Google Search: sharding, moving the entire index into memory in 2001, softening query semantics pre-LLMs, and why retrieval pipelines already resemble modern LLM systems* Pareto frontier strategy: why you need both frontier “Pro” models and low-latency “Flash” models, and how distillation lets smaller models surpass prior generations* Distillation deep dive: ensembles → compression → logits as soft supervision, and why you need the biggest model to make the smallest one good* Latency as a first-class objective: why 10–50x lower latency changes UX entirely, and how future reasoning workloads will demand 10,000 tokens/sec* Energy-based thinking: picojoules per bit, why moving data costs 1000x more than a multiply, batching through the lens of energy, and speculative decoding as amortization* TPU co-design: predicting ML workloads 2–6 years out, speculative hardware features, precision reduction, sparsity, and the constant feedback loop between model architecture and silicon* Sparse models and “outrageously large” networks: trillions of parameters with 1–5% activation, and why sparsity was always the right abstraction* Unified vs. specialized models: abandoning symbolic systems, why general multimodal models tend to dominate vertical silos, and when vertical fine-tuning still makes sense* Long context and the illusion of scale: beyond needle-in-a-haystack benchmarks toward systems that narrow trillions of tokens to 117 relevant documents* Personalized AI: attending to your emails, photos, and documents (with permission), and why retrieval + reasoning will unlock deeply personal assistants* Coding agents: 50 AI interns, crisp specifications as a new core skill, and how ultra-low latency will reshape human–agent collaboration* Why ideas still matter: transformers, sparsity, RL, hardware, systems — scaling wasn't blind; the pieces had to multiply togetherShow Notes:* Gemma 3 Paper* Gemma 3* Gemini 2.5 Report* Jeff Dean's “Software Engineering Advice fromBuilding Large-Scale Distributed Systems” Presentation (with Back of the Envelope Calculations)* Latency Numbers Every Programmer Should Know by Jeff Dean* The Jeff Dean Facts* Jeff Dean Google Bio* Jeff Dean on “Important AI Trends” @Stanford AI Club* Jeff Dean & Noam Shazeer — 25 years at Google (Dwarkesh)—Jeff Dean* LinkedIn: https://www.linkedin.com/in/jeff-dean-8b212555* X: https://x.com/jeffdeanGoogle* https://google.com* https://deepmind.googleFull Video EpisodeTimestamps00:00:04 — Introduction: Alessio & Swyx welcome Jeff Dean, chief AI scientist at Google, to the Latent Space podcast00:00:30 — Owning the Pareto Frontier & balancing frontier vs low-latency models00:01:31 — Frontier models vs Flash models + role of distillation00:03:52 — History of distillation and its original motivation00:05:09 — Distillation's role in modern model scaling00:07:02 — Model hierarchy (Flash, Pro, Ultra) and distillation sources00:07:46 — Flash model economics & wide deployment00:08:10 — Latency importance for complex tasks00:09:19 — Saturation of some tasks and future frontier tasks00:11:26 — On benchmarks, public vs internal00:12:53 — Example long-context benchmarks & limitations00:15:01 — Long-context goals: attending to trillions of tokens00:16:26 — Realistic use cases beyond pure language00:18:04 — Multimodal reasoning and non-text modalities00:19:05 — Importance of vision & motion modalities00:20:11 — Video understanding example (extracting structured info)00:20:47 — Search ranking analogy for LLM retrieval00:23:08 — LLM representations vs keyword search00:24:06 — Early Google search evolution & in-memory index00:26:47 — Design principles for scalable systems00:28:55 — Real-time index updates & recrawl strategies00:30:06 — Classic “Latency numbers every programmer should know”00:32:09 — Cost of memory vs compute and energy emphasis00:34:33 — TPUs & hardware trade-offs for serving models00:35:57 — TPU design decisions & co-design with ML00:38:06 — Adapting model architecture to hardware00:39:50 — Alternatives: energy-based models, speculative decoding00:42:21 — Open research directions: complex workflows, RL00:44:56 — Non-verifiable RL domains & model evaluation00:46:13 — Transition away from symbolic systems toward unified LLMs00:47:59 — Unified models vs specialized ones00:50:38 — Knowledge vs reasoning & retrieval + reasoning00:52:24 — Vertical model specialization & modules00:55:21 — Token count considerations for vertical domains00:56:09 — Low resource languages & contextual learning00:59:22 — Origins: Dean's early neural network work01:10:07 — AI for coding & human–model interaction styles01:15:52 — Importance of crisp specification for coding agents01:19:23 — Prediction: personalized models & state retrieval01:22:36 — Token-per-second targets (10k+) and reasoning throughput01:23:20 — Episode conclusion and thanksTranscriptAlessio Fanelli [00:00:04]: Hey everyone, welcome to the Latent Space podcast. This is Alessio, founder of Kernel Labs, and I'm joined by Swyx, editor of Latent Space. Shawn Wang [00:00:11]: Hello, hello. We're here in the studio with Jeff Dean, chief AI scientist at Google. Welcome. Thanks for having me. It's a bit surreal to have you in the studio. I've watched so many of your talks, and obviously your career has been super legendary. So, I mean, congrats. I think the first thing must be said, congrats on owning the Pareto Frontier.Jeff Dean [00:00:30]: Thank you, thank you. Pareto Frontiers are good. It's good to be out there.Shawn Wang [00:00:34]: Yeah, I mean, I think it's a combination of both. You have to own the Pareto Frontier. You have to have like frontier capability, but also efficiency, and then offer that range of models that people like to use. And, you know, some part of this was started because of your hardware work. Some part of that is your model work, and I'm sure there's lots of secret sauce that you guys have worked on cumulatively. But, like, it's really impressive to see it all come together in, like, this slittily advanced.Jeff Dean [00:01:04]: Yeah, yeah. I mean, I think, as you say, it's not just one thing. It's like a whole bunch of things up and down the stack. And, you know, all of those really combine to help make UNOS able to make highly capable large models, as well as, you know, software techniques to get those large model capabilities into much smaller, lighter weight models that are, you know, much more cost effective and lower latency, but still, you know, quite capable for their size. Yeah.Alessio Fanelli [00:01:31]: How much pressure do you have on, like, having the lower bound of the Pareto Frontier, too? I think, like, the new labs are always trying to push the top performance frontier because they need to raise more money and all of that. And you guys have billions of users. And I think initially when you worked on the CPU, you were thinking about, you know, if everybody that used Google, we use the voice model for, like, three minutes a day, they were like, you need to double your CPU number. Like, what's that discussion today at Google? Like, how do you prioritize frontier versus, like, we have to do this? How do we actually need to deploy it if we build it?Jeff Dean [00:02:03]: Yeah, I mean, I think we always want to have models that are at the frontier or pushing the frontier because I think that's where you see what capabilities now exist that didn't exist at the sort of slightly less capable last year's version or last six months ago version. At the same time, you know, we know those are going to be really useful for a bunch of use cases, but they're going to be a bit slower and a bit more expensive than people might like for a bunch of other broader models. So I think what we want to do is always have kind of a highly capable sort of affordable model that enables a whole bunch of, you know, lower latency use cases. People can use them for agentic coding much more readily and then have the high-end, you know, frontier model that is really useful for, you know, deep reasoning, you know, solving really complicated math problems, those kinds of things. And it's not that. One or the other is useful. They're both useful. So I think we'd like to do both. And also, you know, through distillation, which is a key technique for making the smaller models more capable, you know, you have to have the frontier model in order to then distill it into your smaller model. So it's not like an either or choice. You sort of need that in order to actually get a highly capable, more modest size model. Yeah.Alessio Fanelli [00:03:24]: I mean, you and Jeffrey came up with the solution in 2014.Jeff Dean [00:03:28]: Don't forget, L'Oreal Vinyls as well. Yeah, yeah.Alessio Fanelli [00:03:30]: A long time ago. But like, I'm curious how you think about the cycle of these ideas, even like, you know, sparse models and, you know, how do you reevaluate them? How do you think about in the next generation of model, what is worth revisiting? Like, yeah, they're just kind of like, you know, you worked on so many ideas that end up being influential, but like in the moment, they might not feel that way necessarily. Yeah.Jeff Dean [00:03:52]: I mean, I think distillation was originally motivated because we were seeing that we had a very large image data set at the time, you know, 300 million images that we could train on. And we were seeing that if you create specialists for different subsets of those image categories, you know, this one's going to be really good at sort of mammals, and this one's going to be really good at sort of indoor room scenes or whatever, and you can cluster those categories and train on an enriched stream of data after you do pre-training on a much broader set of images. You get much better performance. If you then treat that whole set of maybe 50 models you've trained as a large ensemble, but that's not a very practical thing to serve, right? So distillation really came about from the idea of, okay, what if we want to actually serve that and train all these independent sort of expert models and then squish it into something that actually fits in a form factor that you can actually serve? And that's, you know, not that different from what we're doing today. You know, often today we're instead of having an ensemble of 50 models. We're having a much larger scale model that we then distill into a much smaller scale model.Shawn Wang [00:05:09]: Yeah. A part of me also wonders if distillation also has a story with the RL revolution. So let me maybe try to articulate what I mean by that, which is you can, RL basically spikes models in a certain part of the distribution. And then you have to sort of, well, you can spike models, but usually sometimes... It might be lossy in other areas and it's kind of like an uneven technique, but you can probably distill it back and you can, I think that the sort of general dream is to be able to advance capabilities without regressing on anything else. And I think like that, that whole capability merging without loss, I feel like it's like, you know, some part of that should be a distillation process, but I can't quite articulate it. I haven't seen much papers about it.Jeff Dean [00:06:01]: Yeah, I mean, I tend to think of one of the key advantages of distillation is that you can have a much smaller model and you can have a very large, you know, training data set and you can get utility out of making many passes over that data set because you're now getting the logits from the much larger model in order to sort of coax the right behavior out of the smaller model that you wouldn't otherwise get with just the hard labels. And so, you know, I think that's what we've observed. Is you can get, you know, very close to your largest model performance with distillation approaches. And that seems to be, you know, a nice sweet spot for a lot of people because it enables us to kind of, for multiple Gemini generations now, we've been able to make the sort of flash version of the next generation as good or even substantially better than the previous generations pro. And I think we're going to keep trying to do that because that seems like a good trend to follow.Shawn Wang [00:07:02]: So, Dara asked, so it was the original map was Flash Pro and Ultra. Are you just sitting on Ultra and distilling from that? Is that like the mother load?Jeff Dean [00:07:12]: I mean, we have a lot of different kinds of models. Some are internal ones that are not necessarily meant to be released or served. Some are, you know, our pro scale model and we can distill from that as well into our Flash scale model. So I think, you know, it's an important set of capabilities to have and also inference time scaling. It can also be a useful thing to improve the capabilities of the model.Shawn Wang [00:07:35]: And yeah, yeah, cool. Yeah. And obviously, I think the economy of Flash is what led to the total dominance. I think the latest number is like 50 trillion tokens. I don't know. I mean, obviously, it's changing every day.Jeff Dean [00:07:46]: Yeah, yeah. But, you know, by market share, hopefully up.Shawn Wang [00:07:50]: No, I mean, there's no I mean, there's just the economics wise, like because Flash is so economical, like you can use it for everything. Like it's in Gmail now. It's in YouTube. Like it's yeah. It's in everything.Jeff Dean [00:08:02]: We're using it more in our search products of various AI mode reviews.Shawn Wang [00:08:05]: Oh, my God. Flash past the AI mode. Oh, my God. Yeah, that's yeah, I didn't even think about that.Jeff Dean [00:08:10]: I mean, I think one of the things that is quite nice about the Flash model is not only is it more affordable, it's also a lower latency. And I think latency is actually a pretty important characteristic for these models because we're going to want models to do much more complicated things that are going to involve, you know, generating many more tokens from when you ask the model to do so. So, you know, if you're going to ask the model to do something until it actually finishes what you ask it to do, because you're going to ask now, not just write me a for loop, but like write me a whole software package to do X or Y or Z. And so having low latency systems that can do that seems really important. And Flash is one direction, one way of doing that. You know, obviously our hardware platforms enable a bunch of interesting aspects of our, you know, serving stack as well, like TPUs, the interconnect between. Chips on the TPUs is actually quite, quite high performance and quite amenable to, for example, long context kind of attention operations, you know, having sparse models with lots of experts. These kinds of things really, really matter a lot in terms of how do you make them servable at scale.Alessio Fanelli [00:09:19]: Yeah. Does it feel like there's some breaking point for like the proto Flash distillation, kind of like one generation delayed? I almost think about almost like the capability as a. In certain tasks, like the pro model today is a saturated, some sort of task. So next generation, that same task will be saturated at the Flash price point. And I think for most of the things that people use models for at some point, the Flash model in two generation will be able to do basically everything. And how do you make it economical to like keep pushing the pro frontier when a lot of the population will be okay with the Flash model? I'm curious how you think about that.Jeff Dean [00:09:59]: I mean, I think that's true. If your distribution of what people are asking people, the models to do is stationary, right? But I think what often happens is as the models become more capable, people ask them to do more, right? So, I mean, I think this happens in my own usage. Like I used to try our models a year ago for some sort of coding task, and it was okay at some simpler things, but wouldn't do work very well for more complicated things. And since then, we've improved dramatically on the more complicated coding tasks. And now I'll ask it to do much more complicated things. And I think that's true, not just of coding, but of, you know, now, you know, can you analyze all the, you know, renewable energy deployments in the world and give me a report on solar panel deployment or whatever. That's a very complicated, you know, more complicated task than people would have asked a year ago. And so you are going to want more capable models to push the frontier in the absence of what people ask the models to do. And that also then gives us. Insight into, okay, where does the, where do things break down? How can we improve the model in these, these particular areas, uh, in order to sort of, um, make the next generation even better.Alessio Fanelli [00:11:11]: Yeah. Are there any benchmarks or like test sets they use internally? Because it's almost like the same benchmarks get reported every time. And it's like, all right, it's like 99 instead of 97. Like, how do you have to keep pushing the team internally to it? Or like, this is what we're building towards. Yeah.Jeff Dean [00:11:26]: I mean, I think. Benchmarks, particularly external ones that are publicly available. Have their utility, but they often kind of have a lifespan of utility where they're introduced and maybe they're quite hard for current models. You know, I, I like to think of the best kinds of benchmarks are ones where the initial scores are like 10 to 20 or 30%, maybe, but not higher. And then you can sort of work on improving that capability for, uh, whatever it is, the benchmark is trying to assess and get it up to like 80, 90%, whatever. I, I think once it hits kind of 95% or something, you get very diminishing returns from really focusing on that benchmark, cuz it's sort of, it's either the case that you've now achieved that capability, or there's also the issue of leakage in public data or very related kind of data being, being in your training data. Um, so we have a bunch of held out internal benchmarks that we really look at where we know that wasn't represented in the training data at all. There are capabilities that we want the model to have. Um, yeah. Yeah. Um, that it doesn't have now, and then we can work on, you know, assessing, you know, how do we make the model better at these kinds of things? Is it, we need different kind of data to train on that's more specialized for this particular kind of task. Do we need, um, you know, a bunch of, uh, you know, architectural improvements or some sort of, uh, model capability improvements, you know, what would help make that better?Shawn Wang [00:12:53]: Is there, is there such an example that you, uh, a benchmark inspired in architectural improvement? Like, uh, I'm just kind of. Jumping on that because you just.Jeff Dean [00:13:02]: Uh, I mean, I think some of the long context capability of the, of the Gemini models that came, I guess, first in 1.5 really were about looking at, okay, we want to have, um, you know,Shawn Wang [00:13:15]: immediately everyone jumped to like completely green charts of like, everyone had, I was like, how did everyone crack this at the same time? Right. Yeah. Yeah.Jeff Dean [00:13:23]: I mean, I think, um, and once you're set, I mean, as you say that needed single needle and a half. Hey, stack benchmark is really saturated for at least context links up to 1, 2 and K or something. Don't actually have, you know, much larger than 1, 2 and 8 K these days or two or something. We're trying to push the frontier of 1 million or 2 million context, which is good because I think there are a lot of use cases where. Yeah. You know, putting a thousand pages of text or putting, you know, multiple hour long videos and the context and then actually being able to make use of that as useful. Try to, to explore the über graduation are fairly large. But the single needle in a haystack benchmark is sort of saturated. So you really want more complicated, sort of multi-needle or more realistic, take all this content and produce this kind of answer from a long context that sort of better assesses what it is people really want to do with long context. Which is not just, you know, can you tell me the product number for this particular thing?Shawn Wang [00:14:31]: Yeah, it's retrieval. It's retrieval within machine learning. It's interesting because I think the more meta level I'm trying to operate at here is you have a benchmark. You're like, okay, I see the architectural thing I need to do in order to go fix that. But should you do it? Because sometimes that's an inductive bias, basically. It's what Jason Wei, who used to work at Google, would say. Exactly the kind of thing. Yeah, you're going to win. Short term. Longer term, I don't know if that's going to scale. You might have to undo that.Jeff Dean [00:15:01]: I mean, I like to sort of not focus on exactly what solution we're going to derive, but what capability would you want? And I think we're very convinced that, you know, long context is useful, but it's way too short today. Right? Like, I think what you would really want is, can I attend to the internet while I answer my question? Right? But that's not going to happen. I think that's going to be solved by purely scaling the existing solutions, which are quadratic. So a million tokens kind of pushes what you can do. You're not going to do that to a trillion tokens, let alone, you know, a billion tokens, let alone a trillion. But I think if you could give the illusion that you can attend to trillions of tokens, that would be amazing. You'd find all kinds of uses for that. You would have attend to the internet. You could attend to the pixels of YouTube and the sort of deeper representations that we can find. You could attend to the form for a single video, but across many videos, you know, on a personal Gemini level, you could attend to all of your personal state with your permission. So like your emails, your photos, your docs, your plane tickets you have. I think that would be really, really useful. And the question is, how do you get algorithmic improvements and system level improvements that get you to something where you actually can attend to trillions of tokens? Right. In a meaningful way. Yeah.Shawn Wang [00:16:26]: But by the way, I think I did some math and it's like, if you spoke all day, every day for eight hours a day, you only generate a maximum of like a hundred K tokens, which like very comfortably fits.Jeff Dean [00:16:38]: Right. But if you then say, okay, I want to be able to understand everything people are putting on videos.Shawn Wang [00:16:46]: Well, also, I think that the classic example is you start going beyond language into like proteins and whatever else is extremely information dense. Yeah. Yeah.Jeff Dean [00:16:55]: I mean, I think one of the things about Gemini's multimodal aspects is we've always wanted it to be multimodal from the start. And so, you know, that sometimes to people means text and images and video sort of human-like and audio, audio, human-like modalities. But I think it's also really useful to have Gemini know about non-human modalities. Yeah. Like LIDAR sensor data from. Yes. Say, Waymo vehicles or. Like robots or, you know, various kinds of health modalities, x-rays and MRIs and imaging and genomics information. And I think there's probably hundreds of modalities of data where you'd like the model to be able to at least be exposed to the fact that this is an interesting modality and has certain meaning in the world. Where even if you haven't trained on all the LIDAR data or MRI data, you could have, because maybe that's not, you know, it doesn't make sense in terms of trade-offs of. You know, what you include in your main pre-training data mix, at least including a little bit of it is actually quite useful. Yeah. Because it sort of tempts the model that this is a thing.Shawn Wang [00:18:04]: Yeah. Do you believe, I mean, since we're on this topic and something I just get to ask you all the questions I always wanted to ask, which is fantastic. Like, are there some king modalities, like modalities that supersede all the other modalities? So a simple example was Vision can, on a pixel level, encode text. And DeepSeq had this DeepSeq CR paper that did that. Vision. And Vision has also been shown to maybe incorporate audio because you can do audio spectrograms and that's, that's also like a Vision capable thing. Like, so, so maybe Vision is just the king modality and like. Yeah.Jeff Dean [00:18:36]: I mean, Vision and Motion are quite important things, right? Motion. Well, like video as opposed to static images, because I mean, there's a reason evolution has evolved eyes like 23 independent ways, because it's such a useful capability for sensing the world around you, which is really what we want these models to be. So I think the only thing that we can be able to do is interpret the things we're seeing or the things we're paying attention to and then help us in using that information to do things. Yeah.Shawn Wang [00:19:05]: I think motion, you know, I still want to shout out, I think Gemini, still the only native video understanding model that's out there. So I use it for YouTube all the time. Nice.Jeff Dean [00:19:15]: Yeah. Yeah. I mean, it's actually, I think people kind of are not necessarily aware of what the Gemini models can actually do. Yeah. Like I have an example I've used in one of my talks. It had like, it was like a YouTube highlight video of 18 memorable sports moments across the last 20 years or something. So it has like Michael Jordan hitting some jump shot at the end of the finals and, you know, some soccer goals and things like that. And you can literally just give it the video and say, can you please make me a table of what all these different events are? What when the date is when they happened? And a short description. And so you get like now an 18 row table of that information extracted from the video, which is, you know, not something most people think of as like a turn video into sequel like table.Alessio Fanelli [00:20:11]: Has there been any discussion inside of Google of like, you mentioned tending to the whole internet, right? Google, it's almost built because a human cannot tend to the whole internet and you need some sort of ranking to find what you need. Yep. That ranking is like much different for an LLM because you can expect a person to look at maybe the first five, six links in a Google search versus for an LLM. Should you expect to have 20 links that are highly relevant? Like how do you internally figure out, you know, how do we build the AI mode that is like maybe like much broader search and span versus like the more human one? Yeah.Jeff Dean [00:20:47]: I mean, I think even pre-language model based work, you know, our ranking systems would be built to start. I mean, I think even pre-language model based work, you know, our ranking systems would be built to start. With a giant number of web pages in our index, many of them are not relevant. So you identify a subset of them that are relevant with very lightweight kinds of methods. You know, you're down to like 30,000 documents or something. And then you gradually refine that to apply more and more sophisticated algorithms and more and more sophisticated sort of signals of various kinds in order to get down to ultimately what you show, which is, you know, the final 10 results or, you know, 10 results plus. Other kinds of information. And I think an LLM based system is not going to be that dissimilar, right? You're going to attend to trillions of tokens, but you're going to want to identify, you know, what are the 30,000 ish documents that are with the, you know, maybe 30 million interesting tokens. And then how do you go from that into what are the 117 documents I really should be paying attention to in order to carry out the tasks that the user has asked? And I think, you know, you can imagine systems where you have, you know, a lot of highly parallel processing to identify those initial 30,000 candidates, maybe with very lightweight kinds of models. Then you have some system that sort of helps you narrow down from 30,000 to the 117 with maybe a little bit more sophisticated model or set of models. And then maybe the final model is the thing that looks. So the 117 things that might be your most capable model. So I think it has to, it's going to be some system like that, that is really enables you to give the illusion of attending to trillions of tokens. Sort of the way Google search gives you, you know, not the illusion, but you are searching the internet, but you're finding, you know, a very small subset of things that are, that are relevant.Shawn Wang [00:22:47]: Yeah. I often tell a lot of people that are not steeped in like Google search history that, well, you know, like Bert was. Like he was like basically immediately inside of Google search and that improves results a lot, right? Like I don't, I don't have any numbers off the top of my head, but like, I'm sure you guys, that's obviously the most important numbers to Google. Yeah.Jeff Dean [00:23:08]: I mean, I think going to an LLM based representation of text and words and so on enables you to get out of the explicit hard notion of, of particular words having to be on the page, but really getting at the notion of this topic of this page or this page. Paragraph is highly relevant to this query. Yeah.Shawn Wang [00:23:28]: I don't think people understand how much LLMs have taken over all these very high traffic system, very high traffic. Yeah. Like it's Google, it's YouTube. YouTube has this like semantics ID thing where it's just like every token or every item in the vocab is a YouTube video or something that predicts the video using a code book, which is absurd to me for YouTube size.Jeff Dean [00:23:50]: And then most recently GROK also for, for XAI, which is like, yeah. I mean, I'll call out even before LLMs were used extensively in search, we put a lot of emphasis on softening the notion of what the user actually entered into the query.Shawn Wang [00:24:06]: So do you have like a history of like, what's the progression? Oh yeah.Jeff Dean [00:24:09]: I mean, I actually gave a talk in, uh, I guess, uh, web search and data mining conference in 2009, uh, where we never actually published any papers about the origins of Google search, uh, sort of, but we went through sort of four or five or six. generations, four or five or six generations of, uh, redesigning of the search and retrieval system, uh, from about 1999 through 2004 or five. And that talk is really about that evolution. And one of the things that really happened in 2001 was we were sort of working to scale the system in multiple dimensions. So one is we wanted to make our index bigger, so we could retrieve from a larger index, which always helps your quality in general. Uh, because if you don't have the page in your index, you're going to not do well. Um, and then we also needed to scale our capacity because we were, our traffic was growing quite extensively. Um, and so we had, you know, a sharded system where you have more and more shards as the index grows, you have like 30 shards. And then if you want to double the index size, you make 60 shards so that you can bound the latency by which you respond for any particular user query. Um, and then as traffic grows, you add, you add more and more replicas of each of those. And so we eventually did the math that realized that in a data center where we had say 60 shards and, um, you know, 20 copies of each shard, we now had 1200 machines, uh, with disks. And we did the math and we're like, Hey, one copy of that index would actually fit in memory across 1200 machines. So in 2001, we introduced, uh, we put our entire index in memory and what that enabled from a quality perspective was amazing. Um, and so we had more and more replicas of each of those. Before you had to be really careful about, you know, how many different terms you looked at for a query, because every one of them would involve a disk seek on every one of the 60 shards. And so you, as you make your index bigger, that becomes even more inefficient. But once you have the whole index in memory, it's totally fine to have 50 terms you throw into the query from the user's original three or four word query, because now you can add synonyms like restaurant and restaurants and cafe and, uh, you know, things like that. Uh, bistro and all these things. And you can suddenly start, uh, sort of really, uh, getting at the meaning of the word as opposed to the exact semantic form the user typed in. And that was, you know, 2001, very much pre LLM, but really it was about softening the, the strict definition of what the user typed in order to get at the meaning.Alessio Fanelli [00:26:47]: What are like principles that you use to like design the systems, especially when you have, I mean, in 2001, the internet is like. Doubling, tripling every year in size is not like, uh, you know, and I think today you kind of see that with LLMs too, where like every year the jumps in size and like capabilities are just so big. Are there just any, you know, principles that you use to like, think about this? Yeah.Jeff Dean [00:27:08]: I mean, I think, uh, you know, first, whenever you're designing a system, you want to understand what are the sort of design parameters that are going to be most important in designing that, you know? So, you know, how many queries per second do you need to handle? How big is the internet? How big is the index you need to handle? How much data do you need to keep for every document in the index? How are you going to look at it when you retrieve things? Um, what happens if traffic were to double or triple, you know, will that system work well? And I think a good design principle is you're going to want to design a system so that the most important characteristics could scale by like factors of five or 10, but probably not beyond that because often what happens is if you design a system for X. And something suddenly becomes a hundred X, that would enable a very different point in the design space that would not make sense at X. But all of a sudden at a hundred X makes total sense. So like going from a disk space index to a in memory index makes a lot of sense once you have enough traffic, because now you have enough replicas of the sort of state on disk that those machines now actually can hold, uh, you know, a full copy of the, uh, index and memory. Yeah. And that all of a sudden enabled. A completely different design that wouldn't have been practical before. Yeah. Um, so I'm, I'm a big fan of thinking through designs in your head, just kind of playing with the design space a little before you actually do a lot of writing of code. But, you know, as you said, in the early days of Google, we were growing the index, uh, quite extensively. We were growing the update rate of the index. So the update rate actually is the parameter that changed the most. Surprising. So it used to be once a month.Shawn Wang [00:28:55]: Yeah.Jeff Dean [00:28:56]: And then we went to a system that could update any particular page in like sub one minute. Okay.Shawn Wang [00:29:02]: Yeah. Because this is a competitive advantage, right?Jeff Dean [00:29:04]: Because all of a sudden news related queries, you know, if you're, if you've got last month's news index, it's not actually that useful for.Shawn Wang [00:29:11]: News is a special beast. Was there any, like you could have split it onto a separate system.Jeff Dean [00:29:15]: Well, we did. We launched a Google news product, but you also want news related queries that people type into the main index to also be sort of updated.Shawn Wang [00:29:23]: So, yeah, it's interesting. And then you have to like classify whether the page is, you have to decide which pages should be updated and what frequency. Oh yeah.Jeff Dean [00:29:30]: There's a whole like, uh, system behind the scenes that's trying to decide update rates and importance of the pages. So even if the update rate seems low, you might still want to recrawl important pages quite often because, uh, the likelihood they change might be low, but the value of having updated is high.Shawn Wang [00:29:50]: Yeah, yeah, yeah, yeah. Uh, well, you know, yeah. This, uh, you know, mention of latency and, and saving things to this reminds me of one of your classics, which I have to bring up, which is latency numbers. Every programmer should know, uh, was there a, was it just a, just a general story behind that? Did you like just write it down?Jeff Dean [00:30:06]: I mean, this has like sort of eight or 10 different kinds of metrics that are like, how long does a cache mistake? How long does branch mispredict take? How long does a reference domain memory take? How long does it take to send, you know, a packet from the U S to the Netherlands or something? Um,Shawn Wang [00:30:21]: why Netherlands, by the way, or is it, is that because of Chrome?Jeff Dean [00:30:25]: Uh, we had a data center in the Netherlands, um, so, I mean, I think this gets to the point of being able to do the back of the envelope calculations. So these are sort of the raw ingredients of those, and you can use them to say, okay, well, if I need to design a system to do image search and thumb nailing or something of the result page, you know, how, what I do that I could pre-compute the image thumbnails. I could like. Try to thumbnail them on the fly from the larger images. What would that do? How much dis bandwidth than I need? How many des seeks would I do? Um, and you can sort of actually do thought experiments in, you know, 30 seconds or a minute with the sort of, uh, basic, uh, basic numbers at your fingertips. Uh, and then as you sort of build software using higher level libraries, you kind of want to develop the same intuitions for how long does it take to, you know, look up something in this particular kind of.Shawn Wang [00:31:21]: I'll see you next time.Shawn Wang [00:31:51]: Which is a simple byte conversion. That's nothing interesting. I wonder if you have any, if you were to update your...Jeff Dean [00:31:58]: I mean, I think it's really good to think about calculations you're doing in a model, either for training or inference.Jeff Dean [00:32:09]: Often a good way to view that is how much state will you need to bring in from memory, either like on-chip SRAM or HBM from the accelerator. Attached memory or DRAM or over the network. And then how expensive is that data motion relative to the cost of, say, an actual multiply in the matrix multiply unit? And that cost is actually really, really low, right? Because it's order, depending on your precision, I think it's like sub one picodule.Shawn Wang [00:32:50]: Oh, okay. You measure it by energy. Yeah. Yeah.Jeff Dean [00:32:52]: Yeah. I mean, it's all going to be about energy and how do you make the most energy efficient system. And then moving data from the SRAM on the other side of the chip, not even off the off chip, but on the other side of the same chip can be, you know, a thousand picodules. Oh, yeah. And so all of a sudden, this is why your accelerators require batching. Because if you move, like, say, the parameter of a model from SRAM on the, on the chip into the multiplier unit, that's going to cost you a thousand picodules. So you better make use of that, that thing that you moved many, many times with. So that's where the batch dimension comes in. Because all of a sudden, you know, if you have a batch of 256 or something, that's not so bad. But if you have a batch of one, that's really not good.Shawn Wang [00:33:40]: Yeah. Yeah. Right.Jeff Dean [00:33:41]: Because then you paid a thousand picodules in order to do your one picodule multiply.Shawn Wang [00:33:46]: I have never heard an energy-based analysis of batching.Jeff Dean [00:33:50]: Yeah. I mean, that's why people batch. Yeah. Ideally, you'd like to use batch size one because the latency would be great.Shawn Wang [00:33:56]: The best latency.Jeff Dean [00:33:56]: But the energy cost and the compute cost inefficiency that you get is quite large. So, yeah.Shawn Wang [00:34:04]: Is there a similar trick like, like, like you did with, you know, putting everything in memory? Like, you know, I think obviously NVIDIA has caused a lot of waves with betting very hard on SRAM with Grok. I wonder if, like, that's something that you already saw with, with the TPUs, right? Like that, that you had to. Uh, to serve at your scale, uh, you probably sort of saw that coming. Like what, what, what hardware, uh, innovations or insights were formed because of what you're seeing there?Jeff Dean [00:34:33]: Yeah. I mean, I think, you know, TPUs have this nice, uh, sort of regular structure of 2D or 3D meshes with a bunch of chips connected. Yeah. And each one of those has HBM attached. Um, I think for serving some kinds of models, uh, you know, you, you pay a lot higher cost. Uh, and time latency, um, bringing things in from HBM than you do bringing them in from, uh, SRAM on the chip. So if you have a small enough model, you can actually do model parallelism, spread it out over lots of chips and you actually get quite good throughput improvements and latency improvements from doing that. And so you're now sort of striping your smallish scale model over say 16 or 64 chips. Uh, but as if you do that and it all fits in. In SRAM, uh, that can be a big win. So yeah, that's not a surprise, but it is a good technique.Alessio Fanelli [00:35:27]: Yeah. What about the TPU design? Like how much do you decide where the improvements have to go? So like, this is like a good example of like, is there a way to bring the thousand picojoules down to 50? Like, is it worth designing a new chip to do that? The extreme is like when people say, oh, you should burn the model on the ASIC and that's kind of like the most extreme thing. How much of it? Is it worth doing an hardware when things change so quickly? Like what was the internal discussion? Yeah.Jeff Dean [00:35:57]: I mean, we, we have a lot of interaction between say the TPU chip design architecture team and the sort of higher level modeling, uh, experts, because you really want to take advantage of being able to co-design what should future TPUs look like based on where we think the sort of ML research puck is going, uh, in some sense, because, uh, you know, as a hardware designer for ML and in particular, you're trying to design a chip starting today and that design might take two years before it even lands in a data center. And then it has to sort of be a reasonable lifetime of the chip to take you three, four or five years. So you're trying to predict two to six years out where, what ML computations will people want to run two to six years out in a very fast changing field. And so having people with interest. Interesting ML research ideas of things we think will start to work in that timeframe or will be more important in that timeframe, uh, really enables us to then get, you know, interesting hardware features put into, you know, TPU N plus two, where TPU N is what we have today.Shawn Wang [00:37:10]: Oh, the cycle time is plus two.Jeff Dean [00:37:12]: Roughly. Wow. Because, uh, I mean, sometimes you can squeeze some changes into N plus one, but, you know, bigger changes are going to require the chip. Yeah. Design be earlier in its lifetime design process. Um, so whenever we can do that, it's generally good. And sometimes you can put in speculative features that maybe won't cost you much chip area, but if it works out, it would make something, you know, 10 times as fast. And if it doesn't work out, well, you burned a little bit of tiny amount of your chip area on that thing, but it's not that big a deal. Uh, sometimes it's a very big change and we want to be pretty sure this is going to work out. So we'll do like lots of carefulness. Uh, ML experimentation to show us, uh, this is actually the, the way we want to go. Yeah.Alessio Fanelli [00:37:58]: Is there a reverse of like, we already committed to this chip design so we can not take the model architecture that way because it doesn't quite fit?Jeff Dean [00:38:06]: Yeah. I mean, you, you definitely have things where you're going to adapt what the model architecture looks like so that they're efficient on the chips that you're going to have for both training and inference of that, of that, uh, generation of model. So I think it kind of goes both ways. Um, you know, sometimes you can take advantage of, you know, lower precision things that are coming in a future generation. So you can, might train it at that lower precision, even if the current generation doesn't quite do that. Mm.Shawn Wang [00:38:40]: Yeah. How low can we go in precision?Jeff Dean [00:38:43]: Because people are saying like ternary is like, uh, yeah, I mean, I'm a big fan of very low precision because I think that gets, that saves you a tremendous amount of time. Right. Because it's picojoules per bit that you're transferring and reducing the number of bits is a really good way to, to reduce that. Um, you know, I think people have gotten a lot of luck, uh, mileage out of having very low bit precision things, but then having scaling factors that apply to a whole bunch of, uh, those, those weights. Scaling. How does it, how does it, okay.Shawn Wang [00:39:15]: Interesting. You, so low, low precision, but scaled up weights. Yeah. Huh. Yeah. Never considered that. Yeah. Interesting. Uh, w w while we're on this topic, you know, I think there's a lot of, um, uh, this, the concept of precision at all is weird when we're sampling, you know, uh, we just, at the end of this, we're going to have all these like chips that I'll do like very good math. And then we're just going to throw a random number generator at the start. So, I mean, there's a movement towards, uh, energy based, uh, models and processors. I'm just curious if you've, obviously you've thought about it, but like, what's your commentary?Jeff Dean [00:39:50]: Yeah. I mean, I think. There's a bunch of interesting trends though. Energy based models is one, you know, diffusion based models, which don't sort of sequentially decode tokens is another, um, you know, speculative decoding is a way that you can get sort of an equivalent, very small.Shawn Wang [00:40:06]: Draft.Jeff Dean [00:40:07]: Batch factor, uh, for like you predict eight tokens out and that enables you to sort of increase the effective batch size of what you're doing by a factor of eight, even, and then you maybe accept five or six of those tokens. So you get. A five, a five X improvement in the amortization of moving weights, uh, into the multipliers to do the prediction for the, the tokens. So these are all really good techniques and I think it's really good to look at them from the lens of, uh, energy, real energy, not energy based models, um, and, and also latency and throughput, right? If you look at things from that lens, that sort of guides you to. Two solutions that are gonna be, uh, you know, better from, uh, you know, being able to serve larger models or, you know, equivalent size models more cheaply and with lower latency.Shawn Wang [00:41:03]: Yeah. Well, I think, I think I, um, it's appealing intellectually, uh, haven't seen it like really hit the mainstream, but, um, I do think that, uh, there's some poetry in the sense that, uh, you know, we don't have to do, uh, a lot of shenanigans if like we fundamentally. Design it into the hardware. Yeah, yeah.Jeff Dean [00:41:23]: I mean, I think there's still a, there's also sort of the more exotic things like analog based, uh, uh, computing substrates as opposed to digital ones. Uh, I'm, you know, I think those are super interesting cause they can be potentially low power. Uh, but I think you often end up wanting to interface that with digital systems and you end up losing a lot of the power advantages in the digital to analog and analog to digital conversions. You end up doing, uh, at the sort of boundaries. And periphery of that system. Um, I still think there's a tremendous distance we can go from where we are today in terms of energy efficiency with sort of, uh, much better and specialized hardware for the models we care about.Shawn Wang [00:42:05]: Yeah.Alessio Fanelli [00:42:06]: Um, any other interesting research ideas that you've seen, or like maybe things that you cannot pursue a Google that you would be interested in seeing researchers take a step at, I guess you have a lot of researchers. Yeah, I guess you have enough, but our, our research.Jeff Dean [00:42:21]: Our research portfolio is pretty broad. I would say, um, I mean, I think, uh, in terms of research directions, there's a whole bunch of, uh, you know, open problems and how do you make these models reliable and able to do much longer, kind of, uh, more complex tasks that have lots of subtasks. How do you orchestrate, you know, maybe one model that's using other models as tools in order to sort of build, uh, things that can accomplish, uh, you know, much more. Yeah. Significant pieces of work, uh, collectively, then you would ask a single model to do. Um, so that's super interesting. How do you get more verifiable, uh, you know, how do you get RL to work for non-verifiable domains? I think it's a pretty interesting open problem because I think that would broaden out the capabilities of the models, the improvements that you're seeing in both math and coding. Uh, if we could apply those to other less verifiable domains, because we've come up with RL techniques that actually enable us to do that. Uh, effectively, that would, that would really make the models improve quite a lot. I think.Alessio Fanelli [00:43:26]: I'm curious, like when we had Noam Brown on the podcast, he said, um, they already proved you can do it with deep research. Um, you kind of have it with AI mode in a way it's not verifiable. I'm curious if there's any thread that you think is interesting there. Like what is it? Both are like information retrieval of JSON. So I wonder if it's like the retrieval is like the verifiable part. That you can score or what are like, yeah, yeah. How, how would you model that, that problem?Jeff Dean [00:43:55]: Yeah. I mean, I think there are ways of having other models that can evaluate the results of what a first model did, maybe even retrieving. Can you have another model that says, is this things, are these things you retrieved relevant? Or can you rate these 2000 things you retrieved to assess which ones are the 50 most relevant or something? Um, I think those kinds of techniques are actually quite effective. Sometimes I can even be the same model, just prompted differently to be a, you know, a critic as opposed to a, uh, actual retrieval system. Yeah.Shawn Wang [00:44:28]: Um, I do think like there, there is that, that weird cliff where like, it feels like we've done the easy stuff and then now it's, but it always feels like that every year. It's like, oh, like we know, we know, and the next part is super hard and nobody's figured it out. And, uh, exactly with this RLVR thing where like everyone's talking about, well, okay, how do we. the next stage of the non-verifiable stuff. And everyone's like, I don't know, you know, Ellen judge.Jeff Dean [00:44:56]: I mean, I feel like the nice thing about this field is there's lots and lots of smart people thinking about creative solutions to some of the problems that we all see. Uh, because I think everyone sort of sees that the models, you know, are great at some things and they fall down around the edges of those things and, and are not as capable as we'd like in those areas. And then coming up with good techniques and trying those. And seeing which ones actually make a difference is sort of what the whole research aspect of this field is, is pushing forward. And I think that's why it's super interesting. You know, if you think about two years ago, we were struggling with GSM, eight K problems, right? Like, you know, Fred has two rabbits. He gets three more rabbits. How many rabbits does he have? That's a pretty far cry from the kinds of mathematics that the models can, and now you're doing IMO and Erdos problems in pure language. Yeah. Yeah. Pure language. So that is a really, really amazing jump in capabilities in, you know, in a year and a half or something. And I think, um, for other areas, it'd be great if we could make that kind of leap. Uh, and you know, we don't exactly see how to do it for some, some areas, but we do see it for some other areas and we're going to work hard on making that better. Yeah.Shawn Wang [00:46:13]: Yeah.Alessio Fanelli [00:46:14]: Like YouTube thumbnail generation. That would be very helpful. We need that. That would be AGI. We need that.Shawn Wang [00:46:20]: That would be. As far as content creators go.Jeff Dean [00:46:22]: I guess I'm not a YouTube creator, so I don't care that much about that problem, but I guess, uh, many people do.Shawn Wang [00:46:27]: It does. Yeah. It doesn't, it doesn't matter. People do judge books by their covers as it turns out. Um, uh, just to draw a bit on the IMO goal. Um, I'm still not over the fact that a year ago we had alpha proof and alpha geometry and all those things. And then this year we were like, screw that we'll just chuck it into Gemini. Yeah. What's your reflection? Like, I think this, this question about. Like the merger of like symbolic systems and like, and, and LMS, uh, was a very much core belief. And then somewhere along the line, people would just said, Nope, we'll just all do it in the LLM.Jeff Dean [00:47:02]: Yeah. I mean, I think it makes a lot of sense to me because, you know, humans manipulate symbols, but we probably don't have like a symbolic representation in our heads. Right. We have some distributed representation that is neural net, like in some way of lots of different neurons. And activation patterns firing when we see certain things and that enables us to reason and plan and, you know, do chains of thought and, you know, roll them back now that, that approach for solving the problem doesn't seem like it's going to work. I'm going to try this one. And, you know, in a lot of ways we're emulating what we intuitively think, uh, is happening inside real brains in neural net based models. So it never made sense to me to have like completely separate. Uh, discrete, uh, symbolic things, and then a completely different way of, of, uh, you know, thinking about those things.Shawn Wang [00:47:59]: Interesting. Yeah. Uh, I mean, it's maybe seems obvious to you, but it wasn't obvious to me a year ago. Yeah.Jeff Dean [00:48:06]: I mean, I do think like that IMO with, you know, translating to lean and using lean and then the next year and also a specialized geometry model. And then this year switching to a single unified model. That is roughly the production model with a little bit more inference budget, uh, is actually, you know, quite good because it shows you that the capabilities of that general model have improved dramatically and, and now you don't need the specialized model. This is actually sort of very similar to the 2013 to 16 era of machine learning, right? Like it used to be, people would train separate models for lots of different, each different problem, right? I have, I want to recognize street signs and something. So I train a street sign. Recognition recognition model, or I want to, you know, decode speech recognition. I have a speech model, right? I think now the era of unified models that do everything is really upon us. And the question is how well do those models generalize to new things they've never been asked to do and they're getting better and better.Shawn Wang [00:49:10]: And you don't need domain experts. Like one of my, uh, so I interviewed ETA who was on, who was on that team. Uh, and he was like, yeah, I, I don't know how they work. I don't know where the IMO competition was held. I don't know the rules of it. I just trained the models, the training models. Yeah. Yeah. And it's kind of interesting that like people with these, this like universal skill set of just like machine learning, you just give them data and give them enough compute and they can kind of tackle any task, which is the bitter lesson, I guess. I don't know. Yeah.Jeff Dean [00:49:39]: I mean, I think, uh, general models, uh, will win out over specialized ones in most cases.Shawn Wang [00:49:45]: Uh, so I want to push there a bit. I think there's one hole here, which is like, uh. There's this concept of like, uh, maybe capacity of a model, like abstractly a model can only contain the number of bits that it has. And, uh, and so it, you know, God knows like Gemini pro is like one to 10 trillion parameters. We don't know, but, uh, the Gemma models, for example, right? Like a lot of people want like the open source local models that are like that, that, that, and, and, uh, they have some knowledge, which is not necessary, right? Like they can't know everything like, like you have the. The luxury of you have the big model and big model should be able to capable of everything. But like when, when you're distilling and you're going down to the small models, you know, you're actually memorizing things that are not useful. Yeah. And so like, how do we, I guess, do we want to extract that? Can we, can we divorce knowledge from reasoning, you know?Jeff Dean [00:50:38]: Yeah. I mean, I think you do want the model to be most effective at reasoning if it can retrieve things, right? Because having the model devote precious parameter space. To remembering obscure facts that could be looked up is actually not the best use of that parameter space, right? Like you might prefer something that is more generally useful in more settings than this obscure fact that it has. Um, so I think that's always attention at the same time. You also don't want your model to be kind of completely detached from, you know, knowing stuff about the world, right? Like it's probably useful to know how long the golden gate be. Bridges just as a general sense of like how long are bridges, right? And, uh, it should have that kind of knowledge. It maybe doesn't need to know how long some teeny little bridge in some other more obscure part of the world is, but, uh, it does help it to have a fair bit of world knowledge and the bigger your model is, the more you can have. Uh, but I do think combining retrieval with sort of reasoning and making the model really good at doing multiple stages of retrieval. Yeah.Shawn Wang [00:51:49]: And reasoning through the intermediate retrieval results is going to be a, a pretty effective way of making the model seem much more capable, because if you think about, say, a personal Gemini, yeah, right?Jeff Dean [00:52:01]: Like we're not going to train Gemini on my email. Probably we'd rather have a single model that, uh, we can then use and use being able to retrieve from my email as a tool and have the model reason about it and retrieve from my photos or whatever, uh, and then make use of that and have multiple. Um, you know, uh, stages of interaction. that makes sense.Alessio Fanelli [00:52:24]: Do you think the vertical models are like, uh, interesting pursuit? Like when people are like, oh, we're building the best healthcare LLM, we're building the best law LLM, are those kind of like short-term stopgaps or?Jeff Dean [00:52:37]: No, I mean, I think, I think vertical models are interesting. Like you want them to start from a pretty good base model, but then you can sort of, uh, sort of viewing them, view them as enriching the data. Data distribution for that particular vertical domain for healthcare, say, um, we're probably not going to train or for say robotics. We're probably not going to train Gemini on all possible robotics data. We, you could train it on because we want it to have a balanced set of capabilities. Um, so we'll expose it to some robotics data, but if you're trying to build a really, really good robotics model, you're going to want to start with that and then train it on more robotics data. And then maybe that would. It's multilingual translation capability, but improve its robotics capabilities. And we're always making these kind of, uh, you know, trade-offs in the data mix that we train the base Gemini models on. You know, we'd love to include data from 200 more languages and as much data as we have for those languages, but that's going to displace some other capabilities of the model. It won't be as good at, um, you know, Pearl programming, you know, it'll still be good at Python programming. Cause we'll include it. Enough. Of that, but there's other long tail computer languages or coding capabilities that it may suffer on or multi, uh, multimodal reasoning capabilities may suffer. Cause we didn't get to expose it to as much data there, but it's really good at multilingual things. So I, I think some combination of specialized models, maybe more modular models. So it'd be nice to have the capability to have those 200 languages, plus this awesome robotics model, plus this awesome healthcare, uh, module that all can be knitted together to work in concert and called upon in different circumstances. Right? Like if I have a health related thing, then it should enable using this health module in conjunction with the main base model to be even better at those kinds of things. Yeah.Shawn Wang [00:54:36]: Installable knowledge. Yeah.Jeff Dean [00:54:37]: Right.Shawn Wang [00:54:38]: Just download as a, as a package.Jeff Dean [00:54:39]: And some of that installable stuff can come from retrieval, but some of it probably should come from preloaded training on, you know, uh, a hundred billion tokens or a trillion tokens of health data. Yeah.Shawn Wang [00:54:51]: And for listeners, I think, uh, I will highlight the Gemma three end paper where they, there was a little bit of that, I think. Yeah.Alessio Fanelli [00:54:56]: Yeah. I guess the question is like, how many billions of tokens do you need to outpace the frontier model improvements? You know, it's like, if I have to make this model better healthcare and the main. Gemini model is still improving. Do I need 50 billion tokens? Can I do it with a hundred, if I need a trillion healthcare tokens, it's like, they're probably not out there that you don't have, you know, I think that's really like the.Jeff Dean [00:55:21]: Well, I mean, I think healthcare is a particularly challenging domain, so there's a lot of healthcare data that, you know, we don't have access to appropriately, but there's a lot of, you know, uh, healthcare organizations that want to train models on their own data. That is not public healthcare data, uh, not public health. But public healthcare data. Um, so I think there are opportunities there to say, partner with a large healthcare organization and train models for their use that are going to be, you know, more bespoke, but probably, uh, might be better than a general model trained on say, public data. Yeah.Shawn Wang [00:55:58]: Yeah. I, I believe, uh, by the way, also this is like somewhat related to the language conversation. Uh, I think one of your, your favorite examples was you can put a low resource language in the context and it just learns. Yeah.Jeff Dean [00:56:09]: Oh, yeah, I think the example we used was Calamon, which is truly low resource because it's only spoken by, I think 120 people in the world and there's no written text.Shawn Wang [00:56:20]: So, yeah. So you can just do it that way. Just put it in the context. Yeah. Yeah. But I think your whole data set in the context, right.Jeff Dean [00:56:27]: If you, if you take a language like, uh, you know, Somali or something, there is a fair bit of Somali text in the world that, uh, or Ethiopian Amharic or something, um, you know, we probably. Yeah. Are not putting all the data from those languages into the Gemini based training. We put some of it, but if you put more of it, you'll improve the capabilities of those models.Shawn Wang [00:56:49]: Yeah.Jeff Dean [00:56:49]:

Patch Tuesday. Preliminary findings from the European Commission come down on TikTok. Switzerland's military cancels its contract with Palantir. Social engineering leads to payroll fraud. Google hands over extensive personal data on a British student activist. Researchers unearth a global espionage operation called “The Shadow Campaigns.” Notepad's newest features could lead to remote code execution. Our guest is Hazel Cerra, Resident Agent in Charge of the Atlantic City Office for the United States Secret Service. Ring says it's all about dogs, but critics hear the whistle. Remember to leave us a 5-star rating and review in your favorite podcast app. Miss an episode? Sign-up for our daily intelligence roundup, Daily Briefing, and you'll never miss a beat. And be sure to follow CyberWire Daily on LinkedIn. CyberWire Guest Today, we're joined by Hazel Cerra, Resident Agent in Charge of the Atlantic City Office for the United States Secret Service, as she discusses the evolution of the Secret Service's investigative mission—from its early focus on financial crimes such as counterfeit currency and credit card fraud to the growing challenges posed by cryptocurrency-related crime. Selected Reading Microsoft February 2026 Patch Tuesday Fixes 58 Vulnerabilities, Six actively Exploited Flaws (Beyond Machines) Adobe Releases February 2026 Patches for Multiple Products (Beyond Machines) ICS Patch Tuesday: Vulnerabilities Addressed by Siemens, Schneider, Aveva, Phoenix Contact (SecurityWeek) Chipmaker Patch Tuesday: Over 80 Vulnerabilities Addressed by Intel and AMD (SecurityWeek) Commission preliminarily finds TikTok's addictive design in breach of the Digital Services Act (European Commission) Palantir's Swiss Exit Highlights Global Data Sovereignty Challenge (NewsCase) Payroll pirates conned the help desk, stole employee's pay (The Register) Google Fulfilled ICE Subpoena Demanding Student Journalist's Bank and Credit Card Numbers (The Intercept) The Shadow Campaigns: Uncovering Global Espionage (Palo Alto Networks Unit 42) Notepad's new Markdown powers served with a side of RCE (The Register) With Ring, American Consumers Built a Surveillance Dragnet (404 Media) Share your feedback. What do you think about CyberWire Daily? Please take a few minutes to share your thoughts with us by completing our brief listener survey. Thank you for helping us continue to improve our show. Want to hear your company in the show? N2K CyberWire helps you reach the industry's most influential leaders and operators, while building visibility, authority, and connectivity across the cybersecurity community. Learn more at sponsor.thecyberwire.com. The CyberWire is a production of N2K Networks, your source for strategic workforce intelligence. © N2K Networks, Inc. Learn more about your ad choices. Visit megaphone.fm/adchoices

Welcome to Get Up in the Cool: Old Time Music with Cameron DeWhitt and Friends. This week's friend is Sparrow Smith! We recorded this on Monday at my home in Portland, Oregon. Tunes in this episode: Katie Morey (0:46) Jewel of the Blue Ridge (Sparrow Smith original) (10:32) Undone in Sorrow (31:47) Young Sally (Sparrow Smith original) (42:34) Chips and Sauce (Ira Bernstein original) (1:00:42) BONUS TRACK: Tend that Flame (Sparrow Smith original) Follow Sparrow Smith on Instagram [Buy her newest album Carolina Mountains](Carolina Mountains | Sparrow Smith - BandcampBandcamphttps://sparrowsmith.bandcamp.com › album › carolina-...) Follow Resonant Rogues on Instagram Visit Resonant Rogues' website Support Get Up in the Cool on Patreon Send Tax Deductible Donations to Get Up in the Cool through Fracture Atlas Sign up at Pitchfork Banjo for my clawhammer instructional series! Schedule a banjo lesson with Cameron Visit Tall Poppy String Band's website and follow us on Instagram follow Sweeten the Third on Instagram

Graphics processing units (GPUs) have become the most important commodity in the AI boom — and have made Nvidia a multi-trillion dollar company. But the tensor processing unit (TPU) could present itself as competition for the GPU.TPUs are developed by Google specifically for AI workloads. And so far, Anthropic, OpenAI and Meta have reportedly made deals for Google's TPUs.Christopher Miller, historian at Tufts University and author of "Chip War: The Fight for the World's Most Critical Technology," explains what this could mean.

Graphics processing units (GPUs) have become the most important commodity in the AI boom — and have made Nvidia a multi-trillion dollar company. But the tensor processing unit (TPU) could present itself as competition for the GPU.TPUs are developed by Google specifically for AI workloads. And so far, Anthropic, OpenAI and Meta have reportedly made deals for Google's TPUs.Christopher Miller, historian at Tufts University and author of "Chip War: The Fight for the World's Most Critical Technology," explains what this could mean.