Podcasts about genai

Jeffrey Matisoff is CEO & Managing Partner, North America at The Brandtech Group, the world's leading Gen AI marketing company. Over more than two decades, he has helped global brands including IBM, Nestlé, and Verizon navigate major shifts in media, technology, and consumer behavior. Before joining Brandtech in 2021, he served as Global President of WPP's dedicated IBM agency, where he led an 800-person global organization spanning creative, media, technology, and analytics. Since joining Brandtech, Jeff has played a key role in the acquisition and transformation of Jellyfish and the development of the company's agentic media platform, helping marketers harness AI to drive growth, efficiency, and competitive advantage.

Accenture (ACN) is down nearly 60% from its recent highs, scraping the absolute bottom of its 52-week lows after its June 2026 earnings release. A tech advisory giant that powers nearly every major Fortune 500 company is suddenly being priced by Wall Street like a dying legacy software company.The mainstream narrative is simple: Gen-AI will replace consultants, automating custom software code and rendering Accenture completely obsolete.But when institutional panic overrides quantitative reality, long-term dividend growth investors look at the data. In this video, we strip away the emotional headlines, look directly at my personal portfolio exposure where I am actively doubling down, and open up the dashboard to see why this might be one of the most asymmetric value opportunities of the decade.We look under the hood at Accenture's 92/100 Quality Score, its perfect 100/100 financial stability rating, its pristine ROIC, and a historic 5.22% forward dividend yield that is compounding at a double-digit rate.Is Accenture a structural value trap or a generationally mispriced compounder? Let's dive into the hard math.

Generative AI isn't coming to government — it's already here. The question is whether agencies lead the change or chase it. How can federal leaders strategically integrate GenAI into workforce planning? How do you harness its power without losing what makes government work — human judgment, accountability, and trust? Join host Michael J. Keegan explores these questions and more on this edition of the Business of Government Hour.See Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.

Annabelle Gerard est VP of AI and Data Business Insights chez Stellantis, l'un des groupes leaders mondiaux de l'automobile, qui regroupe 14 marques dont Peugeot, Citroën, Fiat, Jeep, Chrysler et emploie 250 000 personnes à travers le monde.On aborde :

In this episode, the fourth in our miniseries covering the mAbs journal article collection on artificial intelligence and machine learning in antibody development, we speak to Andrew Buchanan, Senior Vice President of Discovery at a biotech company currently in stealth mode, and former Principle Scientist at AstraZeneca, about his paper in the collection: How to think about designing smart antibodies in the age of GenAI: integrating biology, technology, and experience.Andrew provides a holistic overview of how AI and machine learning are transforming the design of smart antibodies – the more complex evolution of monoclonal antibodies that can bind multiple receptors and utilize different mechanisms of action. Together, we explore the critical role of establishing robust candidate drug target profiles (CDTPs), the current capabilities and limitations of AI in structural antibody design, and how the simultaneous rise of multi-specificity and AI-driven approaches is reshaping the field.Contents[02:10] Exploring the simultaneous rise of AI and multi-specificity in therapeutic antibody design[04:20] Establishing a candidate drug target profile with AI[06:50] Limitations of AI in the development of a CDTP[08:20] AI in practical therapeutic antibody design[10:45] How industry and academia can work together to overcome current limitations in the use of AI in antibody therapeutic design[13:45] Exciting recent applications of AI in antibody design[16:32] Predictions for the next 5 years of AI in antibody design[18:10] If I could grant you a wish to improve the abilities of AI in antibody development, what would it be? Hosted on Acast. See acast.com/privacy for more information.

Paige Christianson, Wyatt Dillenburg, Ava Knabenbauer, Aubrey Keeping, and Brooklyn Butzin are pre-service teaching candidates at the University of Wisconsin-Stevens Point. In this bonus episode of the Journeys of Teaching podcast, they discuss their exploration and critical evaluation of different AI tools during Spring 2026 and the implications of AI for teaching and learning.Dr. Aaron R. Gierhart is an Assistant Professor of Educational Technology at the University of Wisconsin-Stevens Point and previously taught in the Illinois public schools for 11 years. Visit his ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠LinkTree⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ to connect with him. Thank you to Adam Gierhart for the logo artwork. 

Seema Shah and Stephen Sopko discuss the state of the AI trade and continuing impacts infrastructure buildout will have on markets. Seema explains the trends on user consumption in generative AI apps, which she sees increasing. Stephen talks about the timeline to AI driving growth and how adoption will broaden into the wider economy.======== Schwab Network ========Empowering every investor and trader, every market day.Subscribe 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

Life sciences are at a critical inflection point, where scientific innovation, regulatory demands, and patient expectations converge with advances in data and artificial intelligence, positioning IT as a central driver of faster and more effective drug discovery and clinical development.This week, Dave and Rob continue with part 2 off the Life Sciences mini-series with Dr. Alex Zhavoronkov founder and CEO of Insilico Medicine to exploring how drug discovery and clinical development can become faster and more effective, and the role of AI in that process.  TLDR00:40 – Introduction01:00 – Hang out: Kill Bill Vol. 1 & 2 03:07 – Dig in: Life Sciences mini-series, Part 2 06:43 – Conversation with Dr Alex Zhavoronkov 42:12 – The future of AI in drug discovery and a new paradigm for pharma GuestDr. Alex Zhavoronkov: https://www.linkedin.com/in/zhavoronkov/ HostsDave Chapman:  https://www.linkedin.com/in/chapmandr/Esmee van de Giessen:  https://www.linkedin.com/in/esmeevandegiessen/Rob Kernahan:  https://www.linkedin.com/in/rob-kernahan/ ProductionMarcel van der Burg:  https://www.linkedin.com/in/marcel-vd-burg/Dave Chapman:  https://www.linkedin.com/in/chapmandr/ SoundBen Corbett:  https://www.linkedin.com/in/ben-corbett-3b6a11135/Louis Corbett:   https://www.linkedin.com/in/louis-corbett-087250264/ 'Realities Remixed' is an original podcast from Capgemini

Portals and intranets continue to play a critical role in Knowledge Management—whether you're serving internal teams or delivering value directly to clients. In a GenAI-driven world, having a centralized, well-governed home for playbooks, standards, and trusted knowledge is more important than ever.   In this podcast, we spoke with a KM leader who shared insights and lessons learned from building and launching a mature KM platform. The speaker shared practical insights on organizing knowledge, designing sustainable processes, and making thoughtful UI and experience decisions. They also looked back at lessons learned along the way and explored what ongoing governance and maintenance really look like once the platform is live. Moderator: @Brandie Knox - Principal & Creative Director, Knox Design Strategy Speaker: @Caitlin Gibson - Counsel, Debevoise & Plimpton Recorded on 06-17-2026. 

Zipline Roundtable episode: Building Real-Time ML Systems with Zipline + ChrononJoin the Community: https://go.mlops.community/YTJoinInGet the newsletter: https://go.mlops.community/YTNewsletterMLOps GPU Guide: https://go.mlops.community/gpuguideBig shout-out to ZiplineAI for the collaboration!// AbstractReal-time ML use cases like personalization and risk decisioning come with a unique set of challenges: serving fresh feature values at low latency for inference, generating temporally consistent backfills for training, and building complex chains of on-demand, batch, and streaming transformations. In this roundtable, practitioners from Intuit, CreditKarma, Depop, and OpenAI share how they use Zipline and the OSS Chronon project to solve these challenges and deploy real-time ML use cases in production.// BioGerman KrikorianGerman is a Software Engineer on the Feature Platform team at Credit Karma. Since joining the company during the early development of its recommendation system, they have played a key role in building and scaling the platform over the years. Their work focuses on feature pipelines and the feature store, which serves as critical infrastructure supporting numerous teams and business verticals across the organization.Ben MagyarBen is an engineer at Depop working on ML and data systems. Before Depop, he worked on Search at Etsy. Most of his work is around the infrastructure and operational problems that come with running ML systems at scale.Raj KatakamRaj architects ML Infrastructure at Credit Karma (Intuit). He holds a Master's in Software Engineering from Carnegie Mellon and a B.Tech in EECE from IIT Kharagpur. His interests include ML Infrastructure, Distributed Systems, Real-Time Data Processing, and Generative AI. His current focus is on providing feature engineering platforms, production GenAI infrastructure, vector databases, ML model serving, and MLOps pipelines for fraud detection, personalized recommendations, financial insights, and model explainability.Mick JermsurawongLed Flyte ML training/experimentation at Stripe, and now led Chronon for ML features at OpenAIHosted by Demetrios// Related LinksWebsite: https://zipline.ai/https://chronon.ai/~~~~~~~~ ✌️Connect With Us ✌️ ~~~~~~~Catch all episodes, blogs, newsletters, and more: https://go.mlops.community/TYExploreJoin our Slack community [https://go.mlops.community/slack]Follow us on X/Twitter [@mlopscommunity](https://x.com/mlopscommunity) or [LinkedIn](https://go.mlops.community/linkedin)] Sign up for the next meetup: [https://go.mlops.community/register]MLOps Swag/Merch: [https://shop.mlops.community/]Connect with Demetrios on LinkedIn: /dpbrinkmConnect with German on LinkedIn: /e2zdkwh8cxghydg/Connect with Raj on LinkedIn: /rajkiran2190Connect with Mick on LinkedIn:/mick-jermsurawong/

Subscribe to DTC Newsletter - https://dtcnews.link/signupThis episode, Kemberly Gong, VP of Marketing at Contentful, joins Eric to walk through what some marketing leaders are calling “The Great Content Collapse”, and what marketers can actually do about it.The setup: 60% of Google searches now result in zero click-through, and replaced by GenAI models like AI overviews. LLMs already account for 5% of traffic and climbing. Marketing budgets are flat or shrinking. Companies are flooding consumers with AI slop to hit KPIs. And consumers can smell it. 50% lose trust in a brand when they think the content was written by AI.Explore Contentful: https://www.contentful.com/?utm_source=dtc&utm_medium=podcast&utm_campaign=fy27-q2-global-tl_awareness&utm_content=gcc What you'll learn:What is the "great content collapse" and why traditional content strategy is breakingAEO vs SEO: where they overlap and where they divergeWhy agentic agents prefer structured, query-aligned content with third-party validationHow buyer behavior is changing and what marketing teams can do to stay aheadWhere brands over-rely on AI and how to keep the human voiceThe 30-day content audit for the agentic webThe Pets Deli case: 50% conversion lift from one personalization changeThe Ruggable BFCM case: 7x CTR and 25% conversion lift from personalized hero banners + homepagesHow Bossard scaled its content across 18 languages and 38 countries with AI workflows using personalization softwareWhat On Running does to drive 40% of sales onlinePlus: Kemberly Gong's 30-day content audit checklist for the agentic web.Timestamps:00:00 The Great Content Collapse05:38 AEO vs SEO Explained10:08 Why Personalization Wins in 202613:27 Where AI Actually Helps Marketing Teams22:23 Building Brand Trust Across ChannelsSubscribe to DTC Newsletter - https://dtcnews.link/signupAdvertise on DTC - https://dtcnews.link/advertiseWork with Pilothouse - https://dtcnews.link/pilothouseFollow us on Instagram & Twitter - @dtcnewsletterWatch this interview on YouTube - https://dtcnews.link/video

In this episode, PwC's Moritz Kramer shares practical GenAI use cases in supply chain planning, from decision support and scenario analysis to data harmonization, productivity gains, and guided decision-making with humans still in control.Download the ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠episode transcript⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠===== This week, Moritz Kramer of PwC explains how GenAI is moving supply chain planning from static analysis to real-time, decision-centric support. He covers data readiness, black-box transparency, productivity uplift, and why AI will augment planners rather than replace them. ===== Guest 1: Moritz Kramer, Director Integrated Business & Financial Planning, PwC GmbH WPGMoritz is a Director with over 10 years of experience shaping and delivering business‑led supply chain transformations for international organizations. He works across the full transformation journey—from defining strategic direction to translating vision into execution—covering SCM strategy, operating model innovation, organizational and process design, and the implementation of advanced planning solutions, like SAP. His work is driven by the ambition to build resilient, intelligent, and future‑ready supply chains, leveraging digital and AI technologies where they create clear business impact, while keeping a strong focus on value, scalability, and sustainable change.Host 1: Richard Howells, SAP   ⁠⁠⁠⁠⁠⁠⁠Richard Howells⁠⁠⁠⁠⁠⁠⁠ has been working in the Supply Chain Management and Manufacturing space for over 30 years. He is responsible for driving the thought leadership and awareness of SAP's ERP, Finance, and Supply Chain solutions and is an active writer, podcaster, and thought leader on the topics of supply chain, Industry 4.0, digitization, and sustainability.===== Show Links:PwCSupply Chain Management:  ⁠⁠⁠⁠SAP Supply Chain Management⁠⁠⁠⁠ ⁠⁠⁠⁠SAP Insights: Supply Chain⁠⁠⁠⁠       Follow Us on Social Media : Richard Howells: ⁠⁠⁠LinkedIn⁠⁠⁠, SAP Digital Supply Chain: ⁠⁠⁠LinkedIn⁠⁠⁠    Please give us a like, share, and subscribe to stay up-to-date on future episodes!  ===== Chapters:00:00:00: Intro00:01:30: Guest's Introductions00:03:04: Real value of GenAI in supply chain planning00:04:24: Scenario exploration and natural-language responsiveness00:05:12: Productivity uplift and automation of manual work00:05:53: Bridging business and IT with GenAI00:07:45: GenAI as the “black box” explainer00:09:42: Data readiness for GenAI rollout00:11:24: AI for data harmonization00:13:36: AI replacing or augmenting people00:20:20: Biggest implementation obstacles00:23:46: Future trends for AI and GenAI 00:27:09: What is the Future of Supply Chain?00:28:31: Outro

In this episode, Sam talks with Dev Rishi, GM of AI at Rubrik, about what happens when agents move beyond answering questions and start taking action across tools, systems, and business processes. We explore why the enterprise playbook of static guardrails plus human approval starts to break down in the agent era. Agents are useful because they can plan, call tools, update systems, write code, send messages, and operate across workflows at machine speed, but those same capabilities make them difficult to govern with rules written in advance or approval prompts reviewed one at a time. Dev explains why tool access increases blast radius, why agents can route around controls in surprising ways, and why human-in-the-loop review can become security theater when agents operate at scale. We also discuss what enterprises need instead: better visibility, runtime enforcement, policy-aware governance, agent observability, and recovery mechanisms for when something goes wrong. Along the way, we dig into MCP and tool sprawl, small language models for policy enforcement, defense in depth, agent rewind, and why AI may be needed to help secure AI.

How do you build AI that actually understands you and the work you do? It all starts with having the right context.  We talk with Dropbox staff product manager Noorain Noorani and principal engineer Sean-Michael Lewis about the art of context engineering and how Dropbox connects to all the tools your team needs for work—so you get AI that works wherever you do.  ~ ~ ~  Working Smarter is brought to you by Dropbox. Find, organize, and share your work—all in one place—with context-aware AI from Dropbox. You can listen to more episodes of Working Smarter on Apple Podcasts, Spotify, YouTube, Amazon Music, or wherever you get your podcasts. To read more stories and past interviews, visit workingsmarter.ai This show would not be possible without the talented team at Cosmic Standard: producer Ben Montoya, sound engineer Aja Simpson, technical director Jacob Winik, and executive producer Eliza Smith. Special thanks to our illustrator Fanny Luor, marketing consultant Meggan Ellingboe, and editorial support from Catie Keck. Our theme song was composed by Doug Stuart.  Working Smarter is hosted by Matthew Braga. Thanks for listening!

Busting gen AI fears in the workplace requires education, transparent leadership and early success stories. That's the key take-away message of this episode of the Wise Decision Maker Show, which talks about myth-busting Gen AI fears to create a culture of confidence.This article forms the basis for this episode: https://disasteravoidanceexperts.com/myth-busting-gen-ai-fears-to-create-a-culture-of-confidence/

Why This Episode MattersSabrina Maniscalco is one of the few people in quantum who has lived the full arc: two decades of academic work on open quantum systems and non-Markovian noise at Palermo, Turku, Edinburgh, and Helsinki, followed by founding Algorithmiq with three of her former researchers after an early Qiskit Camp. That trajectory matters now because Algorithmiq just had a landmark stretch — sole winner of the $2M Wellcome Leap Q4Bio prize for a quantum-enabled cancer drug discovery workflow, an €18M Series B, a global HQ move to Milan, and its Tensor Network Error Mitigation (TEM) function landing in IBM's Qiskit Functions catalog.If you're trying to make sense of where quantum software actually creates value before fault tolerance arrives — and what a credible "trajectory to advantage" looks like when paired with real clients in life sciences — this is a grounded, technically specific conversation with someone building it.EPISODE SPONSORThis episode is brought to you by Outshift, Cisco's incubation engine. The need for computational power is rapidly increasing in every sector. From drug discovery to material innovation to complex financial modeling, classical systems are reaching their absolute limits. It's time for a paradigm shift. The answer is a scalable quantum network, built on open standards and vendor-agnostic architecture. By uniting distributed quantum devices, you unlock limitless computational power.Learn more about the Cisco Universal Quantum Switch at Outshift.com.Go deeper with the blog post The switch that quantum networking has been waiting for.What We Get IntoWhy a background in open quantum systems and non-Markovian noise turned out to be unusually well-suited to running algorithms on noisy near-term hardwareThe actual science behind the Q4Bio winning workflow: simulating excited-state dynamics of a photosensitizer drug already in Phase II clinical trials, on up to 100 qubitsHow quantum-boosted DMRG works — and why it gives you a built-in benchmark against the best classical method via the bond dimensionThe tradeoff Sabrina would and wouldn't make between more qubits and lower noise, and why neutral atoms' slower sampling rates matter for chemistryWhy even fault-tolerant algorithms like quantum phase estimation still depend on getting state initialization and measurement rightAlgorithmiq's two-product structure: the Digital Quantum Interface (hardware-agnostic infrastructure) and the life sciences application frameworkHow methods built for chemistry are now opening doors into optimization and GenAI — and why that direction emerged from the work, not from a strategy deckWhat the move from Helsinki to Milan signals about the European quantum ecosystem and Algorithmiq's commercial scale-upHow an active learning pipeline is already proposing novel drug variants for synthesis in Prof. Sherri McFarland's labResources & LinksGuest & CompanyAlgorithmiq — The company Sabrina co-founded with Guillermo García-Pérez, Matteo Rossi, and Boris Sokolov; quantum software for life sciences and chemistry.Sabrina Maniscalco — University of Helsinki Research Portal — Publication record covering open quantum systems, non-Markovian dynamics, and quantum information.Sabrina Maniscalco — AI for Good Bio — Consolidated bio covering academic roles and advisory positions, including IQOQI Austria and CERN's Quantum Technology Initiative.The Q4Bio WinAlgorithmiq Wins $2M Wellcome Leap Q4Bio Prize — Company announcement detailing the photodynamic therapy workflow.Wellcome Leap — Q4Bio Prize Announcement — Funder's perspective on finalists and criteria.IBM Quantum Blog — Q4Bio Finalists — IBM's account of the workflow and quantum-classical integration.Funding & HQ MoveTech.eu — Algorithmiq's €18M Series B and Milan move — Coverage of Italy's largest quantum VC round to date.Quantum Computing Report — Algorithmiq Relocates to Milan — Strategic context including the Q4Bio win and IBM partnership.EU-Startups coverage — Investor lineup and Italy's National Quantum Strategy framing.Quantum Advantage & ToolingIBM Quantum Blog — The Dawn of Quantum Advantage — Includes Algorithmiq's TEM (Tensor Network Error Mitigation) function in the Qiskit Functions catalog.Algorithmiq & IBM Quantum Advantage Tracker — The heterogeneous materials experiment Algorithmiq and IBM put forward as a community benchmark.Silicon Republic interview with Sabrina — Useful prior context on her philosophy of using quantum to simulate quantum systems.Key Quotes & InsightsOn the foundation of the company's approach: "We learned very early what we thought were the bottlenecks of quantum computers — what you really need to worry about if you want to implement computation at scale." A direct line from Qiskit Camp Vermont to Algorithmiq's product strategy.On Q4Bio, in Sabrina's words: "This molecule is already in Phase II clinical trial. So it's not hydrogen. It's a real molecule." A useful counter to the common critique that quantum chemistry demos still live in toy-model land.On quantum-boosted DMRG (insight): In the worst case, the method matches the best classical technique; in the better case, it outperforms it — and the bond dimension tells you which regime you're in. Built-in benchmarking against the classical baseline.On the hardware tradeoff: Asked whether she'd prefer 100 higher-fidelity qubits or 200 noisier ones, Sabrina's answer is "it depends" — and the explanation about why neutral atoms' lower sampling rates limit chemistry use cases is one of the more concrete things you'll hear on platform tradeoffs.On strategy (insight): New verticals at Algorithmiq are ...

In this episode, we break down the immediate plunge in global oil prices and the stock market rally triggered by the historic US-Iran framework peace agreement. We also look at the logistics at sea as nearly 600 stranded tankers cautiously await the reopening of the Strait of Hormuz, and highlight the UAE's major governance leap with the creation of the federal Artificial Intelligence and Data Authority.

Join Jordan Bayne, Founder of The Squad, the Film3™ brand, and Co-founder of enGEN3, for a radical re-evaluation of how intellectual property operates in an automated world. An acclaimed filmmaker whose Oscar-contender short The Sea Is All I Know starred Melissa Leo and screened at Cannes, Jordan has spent her career on the bleeding edge of narrative expression. Today, as Head of GenAI + Web3 for Goldfinch, she is tackling the ultimate creative crisis: the exploitation of IP by generative models. In this episode, we explore how the trademarked Film3™ movement uses Web3 guardrails and decentralized networks to turn passive audiences into active IP co-owners, creating a massive defensive wall for creators against corporate AI exploitation.

On this week's Crewcast, we recap some of our favourite games from this year's Summer Game Fest, including on-the-ground reporting from Frank Howley. The California Game Dev Problem by Del Walker (this is the video Jeremy mentioned): https://www.youtube.com/watch?v=1bJUSTTUDl8 Frank on the Giant Bomb couch: https://youtu.be/XPKk-cZ0g80?t=3658 iTunes Page: https://itunes.apple.com/us/podcast/noclip/id1385062988 RSS Feed: http://noclippodcast.libsyn.com/rss Spotify: https://open.spotify.com/show/5XYk92ubrXpvPVk1lin4VB?si=JRAcPnlvQ0-YJWU9XiW9pg Watch our docs: https://youtube.com/noclipvideo Crewcast channel: https://www.youtube.com/channel/noclippodcast Follow our games coverage escapades: https://www.youtube.com/@Noclip2 Learn About Noclip: https://www.noclip.video Become a Patron and get early access to new episodes: https://www.patreon.com/noclip Chapters: 0:00:00 - Intro 0:02:19 - Thanking our Patreon supporters! 0:04:00 - Frank Howley's Summer Game Fest 0:23:52 - gen Atlas 0:25:57 - Rhythm Heaven Groove 0:27:40 - Carcass Clad 0:31:15 - Signet City 0:33:36 - Bancho the Chef 0:34:57 - Kemuri 0:36:28 - Dynasty Warriors 3: Complete Edition 0:37:27 - Ace Combat 8: Wings of Theve 0:39:22 - Silent Hill Townfall 0:43:57 - Oh My Doug! 0:45:20 - Neighborhoods 0:48:01 - God of War: Laufey 0:54:36 - Rayman Legends Retold 0:56:59 - TMNT - The Last Ronin 0:58:00 - Stage Tour 0:59:27 - Touhou Koumakyou: New Classic - the Embodiment of Scarlet Devil 1:01:36 - Muramasa: Revenant Blades 1:02:21 - Vivarium 1:07:23 - Mighty Cuphead Adventure 1:14:25 - Virtue and a Sledgehammer 1:14:55 - Janet DeMornay Is A Slumlord (and a Witch) 1:15:43 - The Hearth and Harbour 1:16:27 - Mr. Records 1:17:07 - Bub 1:19:48 - Gears of War: E-Day, Control Resonant, MGS Collection Vol. 2, and Fire Emblem: Fortune's Weave 1:21:19 - The Legend of Zelda: Ocarina of Time Remake 1:36:06 - Stellar Blade: Blood Rain 1:38:36 - Viva Piñata 1:46:55 - Q: How do you feel about AAA secretly using genAI? 1:58:46 - Q: What is the heart of an MMO? Story? Endgame? 2:02:46 - Q: What's the best way to preserve MP experiences? 2:07:01 - Q: What media do you find personally inspiring? 2:20:40 - Sign Off

Successful AI transformation depends not only on technology but also on clear, consistent, multi-channel communication that keeps employees informed, engaged, and involved in the change process. That's the key take-away message of this episode of the Wise Decision Maker Show, which talks about why your employees aren't hearing you on Gen AI transformation.This article forms the basis for this episode: https://disasteravoidanceexperts.com/why-your-employees-arent-hearing-you-on-gen-ai-transformation/

Continuamos con el programa más huevón del mundo, en donde lo único que aportamos es leer los comentarios sobre algunos de los temas o posts más peliagudos recientes, y hoy tocó el turno del villano de moda: la IA. Churros y Palomitas es tu portal a los mejores análisis y cotorreo relacionado con el cine, establecido en 2024 y es uno de los podcasts más longevos en el mundo del entretenimiento. En esta entrega hablamos de las críticas a Martin Scorsese por validar el uso de la tecnología de Black Forest Labs usando IA para crear storyboards así como del anuncio primero de participación y luego de salida de Jorge R. Gutiérrez en el piloto de Gen AI para creadores de Amazon. Esta entrega fue traída gracias a:Productora Ejecutiva: Blanca LópezCo-Productor: Román RangelAgradecimiento especial a nuestros Patreons: Adriana Fernández, Agustín Galván, Cris Mendoza, Fernando de Anda, Franky González, Jaime Rosales, Juan Espíritu, Zert, Álvaro Vázquez, Arturo Manrique, Fabiola Sándoval, Lau Berdejo, Marce, Miguel Huesca, Alejandro Alemán, Arturo Aguilar, Enrique Vázquez, Ernesto Diezmartínez, Mariana Padilla, Tania RG y Fernando Alonso.¡Gracias a nuestros suscriptores en Twitch ! Como el buen Jiff01 y CoyoteraxTú también puedes apoyar la creación de este y más programas y recibir crédito (para que aumentes currículum) y otros extras exclusivos en www.patreon.com/churrosypalomitas¿Quieren continuar la discusión? Tenemos nuestro canal de Discord de Charlas y Palomitas, con distintos temas, unos solo para productores del show y otros para toda la banda.

Innovation isn't about funding, it's about how organisations are built and led. Progress comes from cutting bureaucracy, empowering mission-led teams, and asking the right questions to unlock bold breakthroughs. This week, Dave, Esmee and Rob are joined again by André Loesekrug-Pietri, Chair and Scientific Director of the Joint European Disruptive Initiative (JEDI, Europe's ARPA) to explore how Europe can turn moonshot ambitions into reality by building the right people, culture and operating models for future-shaping organisations. TLDR00:41 – Introduction01:14 – Hang out: Esmee returns and the missing API has been found!05:14 – Dig in: Staying in step with global innovation12:57 – Conversation with André Loesekrug-Pietri1:02:26 – Roland Garros tennis, and unlocking creative energy GuestAndre Loeskrug-Petri: https://www.linkedin.com/in/andrepietri/X: @eurojediwww.jedi.foundation HostsDave Chapman:  https://www.linkedin.com/in/chapmandr/Esmee van de Giessen:  https://www.linkedin.com/in/esmeevandegiessen/Rob Kernahan:  https://www.linkedin.com/in/rob-kernahan/ ProductionMarcel van der Burg:  https://www.linkedin.com/in/marcel-vd-burg/Dave Chapman:  https://www.linkedin.com/in/chapmandr/ SoundBen Corbett:  https://www.linkedin.com/in/ben-corbett-3b6a11135/Louis Corbett:   https://www.linkedin.com/in/louis-corbett-087250264/ 'Realities Remixed' is an original podcast from Capgemini

A mass extinction event approaches, but not because of a meteor or global warming. The culprit this time is AI, and in particular, Large Language Models like ChatGPT and Claude. In their crosshairs are a slew of vendors selling analytical functionality: dashboards, visualizations, analyses, semantic layers, OLAP cubes and the like. For decades, these vendors have dominated enterprise decision-making, commanding 5, 6, even 7-figure pricetags to provide the scaffolding needed for insights. The LLMs now threaten that entire landscape, disrupting the foundation of data-driven workflows. True, the results of GenAI can be inaccurate, but their ease of use and relatively low cost will trump those concerns. Check out this hard-hitting session to hear DM Radio Host, Eric Kavanagh explain what's happening and why it matters to analysts everywhere. He'll examine the impact on the analytics industry, and explore ways that these vendors can stay relevant. He'll also offer advice for businesses looking to remain data-driven, and why data prep is where the action will be.

*Disclaimer: This episode is a little late and some of the events mentioned have already passed.This week on Sludge we Introduce our newest host Rowan, joining Marcel for the first time on the May episode of GroundUp. They dive into the wave of AI generated posters, music, and art that has the Cape Town scene divided. We're also proud to introduce SAHARA, the South African Heavy/Alternative Release Archive, a project Rowan and Meagan have been working on behind the scenes, aimed at documenting all of the latest releases within the alternative and metal space! There's been some incredible release's from Human Nebula, Die Gemeente, BCOM, and more! So catch the scoop on those! Give us your thoughts on AI generated content in artistic spaces.What are some of your favourite releases in the month of May?What are your thoughts on SAHARA?Join the discord to become a part of the discussion:https://discord.gg/JUDEUeKTPK Explore our exclusive merch store for unique and high-quality items inspired by our podcast! From stylish t-shirts to snug beanies, there's something for every listener. Show your support and grab your favourite merch today!Support the showHelp us continue making great content for listeners everywhere by subscribing to Sludge Underground Podcast +Websitehttps://www.sludgeunderground.comMerchhttps://sludgeunderground.store/Instagramhttps://www.instagram.com/sludgeundergroundTikTokhttps://www.tiktok.com/@sludgeundergroundYouTubehttps://www.youtube.com/@sludgeundergroundTwitterhttps://twitter.com/Sludge031Facebookhttps://www.facebook.com/SludgeUnderground

In this Diving Deep episode, Dr. Robert Pearl and Jeremy Corr examine the rapid advance of generative AI, along with the growing conflict between medicine's mission to heal and doctors' need for financial security. The conversation begins with a question now echoing across every profession: Will AI replace highly trained workers? In medicine, Dr. Pearl argues, the answer is less about replacement than redefinition. Drawing on recent changes in software development, he explains how “vibe coding” has allowed programmers to stop writing much of the code themselves and instead use generative AI to build, test and refine applications from plain-language instructions. Rather than feeling diminished, many coders report greater satisfaction because AI has taken over the repetitive, error-prone work and left them more time for problem-solving. Pearl sees a similar possibility in healthcare. Like coding, medicine relies on years of training, structured reasoning and repeatable processes. Chronic disease management offers the clearest example. Hypertension, diabetes and high cholesterol are leading causes of heart attacks, strokes and kidney failure, yet proven treatments often fail because doctors lack the time to monitor patients continuously and adjust medications quickly. With home devices, physician-set targets and generative AI support, care could shift from occasional office visits to ongoing management, helping more patients achieve control while freeing physicians to focus on complex cases. The second half of the episode turns from technology to mission. Using Tim Cook's legacy at Apple as a case study, Pearl examines what happens when values and financial incentives collide. Cook's tenure produced extraordinary business results, but critics have questioned whether some of his choices conflicted with his own values and Apple's public statements around privacy, dignity and human-centered technology. Pearl uses that as background for a similar question about medicine: What happens when doctors, who train to help and heal others above all else, feel increasingly forced to make career decisions shaped by money? For generations, medicine was understood as a calling. Today, most physicians no longer own their practices. Many now work for hospitals, health systems, insurers or private equity-backed groups, while others have moved into concierge or direct primary care models. Pearl stresses that these choices are rational. But the financial upside comes with psychological and moral consequences that are rarely discussed — and that may shape the future of physician fulfillment. For more, tune into this month's episode and check out the links below. Helpful links The AI Revolution In Coding Offers A Preview Of Medicine's Future (Forbes) What Tim Cook's Legacy Teaches Doctors About Money And Mission (Forbes) Monthly Musings on American Healthcare (RobertPearlMD.com) * * * Dr. Robert Pearl is the author of “ChatGPT, MD: How AI-Empowered Patients & Doctors Can Take Back Control of American Medicine.” Fixing Healthcare is a co-production of Dr. Robert Pearl and Jeremy Corr. Subscribe to the show via Apple, Spotify or wherever you find podcasts. Join the conversation or suggest a guest by following the show on X and LinkedIn. The post FHC #217: GenAI, physician fulfillment & the future of medical practice appeared first on Fixing Healthcare.

In episode 191 of Cybersecurity Where You Are, Sean Atkinson sits down with Sasha Elvenaes, Sr. Multidimensional Threat Analyst at the Center for Internet Security® (CIS®), and Rian Davis, Multidimensional Threat Analyst at CIS. Together, they discuss how threat actors are misusing generative artificial intelligence (GenAI) to plan physical threats.Here are some highlights from our episode:00:40. Introductions to Sasha, Rian, and their research on GenAI misuse01:56. The impact of GenAI on lowering the barrier for operationalizing physical threat activity03:37. Exploitation of GenAI model design to circumvent models' guardrails05:58. The misuse of session persistence to streamline physical threat research07:57. GenAI misuse: A call for critical infrastructure operators to think about security differently11:52. Factors that make large-scale events a target of physical threat activity14:33. The use of GenAI as a strategy for organizations to see what threat actors could see15:37. Ongoing question: How can drones help mitigate risks while protecting public safety?17:13. Extrapolation as a reinforcement of GenAI session persistence20:15. The new reality: Look at what information AI can provide to threat actors25:01. Traditional methods vs. GenAI conversations for threat planning27:58. Continuous vulnerability assessments, communication, and other recommendationsResourcesAn Examination of Generative AI and Physical Threat PlanningAn Examination of AI-Enabled Threats to Event and Stadium SecurityMultidimensional ThreatsMan who exploded Cybertruck in Las Vegas used ChatGPT in planning, police sayEpisode 190: Separating Mythos AI Fact from FictionEpisode 185: AI Prompt Injection from a Risk Perspective5 Steps to Help Secure Your City before a Large-Scale EventUnmanned Aircraft Systems (UAS): Evolving Risks to Large-Scale Public Gatherings8 Security Essentials for Managing Your Online PresenceVulnerability AssessmentsIf you have some feedback or an idea for an upcoming episode of Cybersecurity Where You Are, let us know by emailing podcast@cisecurity.org.

AI implementation in higher education is often framed as a technology question. California State University treated it as change management with technology as the catalyst, rolling out ChatGPT Edu to 22 universities in 18 months while running the largest AI survey ever conducted at a single university system. In this episode of the Changing Higher Ed® podcast, Dr. Drumm McNaughton speaks with Dr. Leslie Kennedy, Assistant Vice Chancellor for Academic Technology Services at the California State University Office of the Chancellor, about how the system designed and executed its generative AI implementation and what the Ahead of the Curve survey of 94,060 respondents reveals about AI adoption, faculty engagement, and student behavior. Drawing on her work co-leading the academic side of CSU's GenAI initiative, Kennedy explains the governance structure that made the rollout possible, the campus-level training infrastructure that scaled adoption across 22 universities, and the survey findings that pushed back on common assumptions about cheating, faculty resistance, and AI access gaps. This conversation is especially relevant for presidents, provosts, boards, and CIOs evaluating how to move from AI policy discussions to systemwide implementation. Topics Covered: The sequencing model behind CSU's 18-month AI rollout Findings from the largest AI survey ever conducted at a single university system Why faculty are the only group reporting both positive and negative AI impact How CSU funded faculty-led innovation through the AI Educational Innovations Challenge The communication challenges of running AI implementation across 22 independent campuses What CSU plans next: hackathons, embedded credentials, and domain-specific tools Real-World Examples Discussed: The AI Educational Innovations Challenge received 417 faculty applications against an expected 50, with 63 funded at $3M ChatGPT Edu deployment across all 22 CSU campuses, now at 225,000 active accounts Student hackathons run with IBM Watson, AWS, NVIDIA, and Cal Poly partners across multiple disciplines Faculty-led podcasts (My Robot Teacher from Cal Maritime and Unfixed from Chico State) that built peer-to-peer training resources Three Key Takeaways for Leadership: Sequencing matters more than budget or technology. Faculty resolution first, governance second, enterprise tool third, training and funded experimentation in parallel. Faculty carry more complexity than staff or students in AI implementation, and need different support, training cadence, and communication than other groups. Communication is a continuous operating discipline, not a launch campaign. The technology changes faster than any single training cycle. This episode offers a practical view of what large-scale AI implementation actually looks like in higher education, and why the institutions getting it right are treating it as change management work supported by technology rather than a technology rollout in search of governance. Read the transcript: https://changinghighered.com/https://changinghighered.com/csu-chatgpt-edu-rollout-lessons-higher-ed-leaders/ #GenerativeAI #HigherEducation #HigherEducationPodcast

Hubris, arrogance, dérapage : ces mots dédouanent les patrons de la tech au lieu de les nommer pour ce qu'ils sont.Cette chronique dévoile ce que la presse n'ose pas dire : ce que Musk, Zuckerberg et leurs pairs exercent sur nos démocraties, c'est une violence structurelle... et la résistance commence par l'appeler ainsi.On refait la Tech, la chronique socio-historique de Trench Tech animée par Gérald Holubowicz, en collaboration avec Synth. ***** À PROPOS DE TRENCH TECH *****LE talkshow « Esprits Critiques pour Tech Ethique »Écoutez-nous sur toutes les plateformes de podcast

This week on Marketing O'Clock: Google's new Search Generative AI Performance Reports are starting to head to Search Console; while limited to certain UK-based sites now, this feature will eventually be global. Maggie Humphrey - PMAX vs Standard Shopping: Driving Ecommerce Growth in Google Ads-https://speakerdeck.com/mhumprey/pmax-vs-standard-shopping-driving-ecommerce-growth-in-google-ads-smx-advanced-2026Plus, hear from some of the leading voices in the marketing world who were present at SMX Advanced! Visit us at - https://marketingoclock.com/

In this talk, Nikita, Senior Applied Data Scientist at the AWS Generative AI Innovation Center, shares his expertise in bringing enterprise artificial intelligence out of the sandbox—from his early days optimizing traditional machine learning models like gradient boosting to deploying advanced production-grade GenAI pipelines. We explore what it really takes to move generative AI systems from pilot prototypes to production environments.Links:- AWS Generative AI Innovation Center: https://aws.amazon.com/ai/generative-ai/innovation-center/You'll learn about:- Deploying multi-layered defenses independent of backend LLMs.- Evaluating parameter-efficient methods like LoRA and QLoRA for small models.- Balancing long-term domain expertise with real-time documentation retrieval.- Utilizing multi-agent orchestration for search and anomaly explanation.- Setting up robust LLM-as-a-judge frameworks verified by human metrics.- Leveraging Amazon Bedrock components for memory and runtime scalability.TIMECODES:05:52 Shifting from traditional ML to generative AI07:49 Hybrid pipelines blending classical ML and LLMs11:25 Production guardrails and multi-layered system defense16:15 Prompt bypasses, input attacks, and AI red teaming20:49 Newsletter localization and translation with Zalando27:24 Evaluation frameworks and human-in-the-loop metrics33:07 Aligning LLM-as-a-judge with few-shot prompts34:49 Fine-tuning small language models versus prompting41:18 Complementary mechanics of RAG and fine-tuning43:00 Agentic web search tools for anomaly explanation47:01 Automated text generation from real-time sports sensors49:58 AWS project scoping and proof of concept timelines54:58 Interview requirements and career skills for AWS roles57:59 Enterprise architecture patterns and system observability01:00:42 Reusable infrastructure blocks on Amazon BedrockThis session is designed for machine learning engineers, data scientists, and technical product managers looking to architect reliable, production-ready GenAI workflows. It is highly valuable for teams aiming to bridge the gap between experimental AI prototypes and secure enterprise software.Connect with DataTalks.Club:- Join the community - https://datatalks.club/slack.html- Subscribe to our Google calendar to have all our events in your calendar - https://calendar.google.com/calendar/r?cid=ZjhxaWRqbnEwamhzY3A4ODA5azFlZ2hzNjBAZ3JvdXAuY2FsZW5kYXIuZ29vZ2xlLmNvbQ- Check other upcoming events - https://lu.ma/dtc-events- GitHub: https://github.com/DataTalksClub- LinkedIn - https://www.linkedin.com/company/datatalks-club/ - Twitter - https://twitter.com/DataTalksClub - Website - https://datatalks.club/ Connect with Nikita- Linkedin - https://www.linkedin.com/in/kozodoi/- Github - https://github.com/kozodoi- Website and blog - https://www.kozodoi.me/

Is the industry sleepwalking into a creative quality crisis? Why are we still building retail media on first-wave infrastructure? Are marketers ready to adopt audio as a full funnel channel?This week, we're handing the mic over to our audience for another mailbag special. We've gathered burning questions from across the ad tech landscape and put them to our team, who'll dive in and answer as best they can… In this episode of The MadTech Podcast, ExchangeWire's head of content, John Still, is joined by CEO Rachel Smith and COO Lindsay Rowntree to discuss generative AI content, retail media, and audio. The conversation explores if AI is doing more harm than good to public perceptions of advertising, the pace of retail media's evolution, and measurement challenges with audio channels.Thank you to DAIVID's Barney Worfolk-Smith, VML's Kiessé Lamour, and Audion's Kamel El Hadef for this month's questions!Got a burning question for the ExchangeWire team? Send your questions over to us and we'll answer in our next monthly mailbag special...0:00 Introduction0:47 Is the industry sleepwalking into a creative quality crisis?20:46 Why are we still building retail media on first-wave infrastructure?33:53 Are marketers ready to adopt audio as a full funnel channel?

Realities Remixed, formerly known as Cloud Realities, launches a new season exploring the intersection of people, culture, industry and tech.Life sciences are at a turning point, where scientific innovation, regulatory pressure, and patient expectations collide with unprecedented advances in data, AI, and digital platforms. IT is no longer a supporting function but a critical driver of how therapies are discovered, developed, scaled, and delivered safely and at speed.This week, Dave and Rob kick off the Life Sciences mini‑series with Thorsten Rall, Global Industry Lead for Life Sciences at Capgemini, to exploring the current state of the sector, the key themes shaping the episodes ahead, and what it takes to drive better patient outcomes. TLDR00:30 – Introduction to Life Sciences and co‑host Thorsten Rall04:37 – Hang‑out: Navigating Waterloo Station07:50 – Deep dive with Thorsten Rall into the Life Sciences landscape28:03 - What are the main challenges in the sector and main themes45:31 – BBQ season is starting HostsDave Chapman:  https://www.linkedin.com/in/chapmandr/Esmee van de Giessen:  https://www.linkedin.com/in/esmeevandegiessen/Rob Kernahan:  https://www.linkedin.com/in/rob-kernahan/with co-host Thorsten Rall: https://www.linkedin.com/in/thorsten-alexander-rall-b232185/ ProductionMarcel van der Burg:  https://www.linkedin.com/in/marcel-vd-burg/Dave Chapman:  https://www.linkedin.com/in/chapmandr/ SoundBen Corbett:  https://www.linkedin.com/in/ben-corbett-3b6a11135/Louis Corbett:   https://www.linkedin.com/in/louis-corbett-087250264/ 'Realities Remixed' is an original podcast from Capgemini

Modern work can be frustrating and chaotic—if you don't have the right tools. From context engineering to multimodal search, go behind the scenes and hear how Dropbox engineers are building AI that actually understands you, so you can focus on the work that matters most. If you're new to Working Smarter, we've travelled from the F1 track to the bottom of a lake, and heard real stories from chefs, doctors, lawyers, and founders about how AI is helping them do more of what they love about their jobs. But in our third season, we're talking to the people behind the tools—the engineers and product leaders building helpful, time-saving AI features into the Dropbox experience you already know and trust. You'll hear all about their work on agents, inference, security, and, of course, how the people building AI use AI themselves. ~ ~ ~  Working Smarter is brought to you by Dropbox. Find, organize, and share your work—all in one place—with context-aware AI from Dropbox. You can listen to more episodes of Working Smarter on Apple Podcasts, Spotify, YouTube, Amazon Music, or wherever you get your podcasts. To read more stories and past interviews, visit workingsmarter.ai This show would not be possible without the talented team at Cosmic Standard: producer Ben Montoya, sound engineer Aja Simpson, technical director Jacob Winik, and executive producer Eliza Smith. Special thanks to our illustrator Fanny Luor, marketing consultant Meggan Ellingboe, and editorial support from Catie Keck.  Our theme song was composed by Doug Stuart.  Working Smarter is hosted by Matthew Braga. Thanks for listening!

Twenty-five years ago, the goal was to build a website as a digital "single source of truth." In an era of AI agents and hyper-personalized realities, is the very concept of a single, universal brand "truth" now an obstacle to creating a truly relevant customer experience?Agility requires not just adopting new channels and technologies, but fundamentally rethinking the role of content and data in a constantly shifting landscape. It's the ability to move from managing a static digital property to orchestrating a fluid, dynamic relationship with your audience.Today, we're going to talk about the 25-year evolution of digital experience, from the early days of enterprise content management to today's complex ecosystem of AI-driven, composable platforms. We'll explore how seismic shifts—from the introduction of the iPhone to the rise of agentic AI—have not just changed the tools we use, but have fundamentally redefined the relationship between brands and their customers.To help me discuss this topic, I'd like to welcome, Michelle Boockoff-Bajdek, CMO at Sitecore, a company that is turning 25 this year and has managed to maintain its leadership in the space through many changes and a few curveballs. About Michelle Boockoff-Bajdek Michelle Boockoff-Bajdek (BB) is the Chief Marketing Officer at Sitecore, where she leads a global team of marketers who are redefining what's possible in modern marketing. Together, they're putting the power of generative and agentic AI to work – creating digital experiences that connect people and possibilities across the globe. Michelle is a trailblazer in bringing AI into marketing. At IBM, she served first as Global Head of B2B Marketing at the Weather Company, and then as CMO of IBM Watson, the company's pioneering AI platform. There, she served as the steward of the Watson brand, helping the world understand how AI can transform both work and life. Since then, she's become a Fellow at the Marketing Academy, a founding member of CMO Huddles, and has held CMO roles at both Skillsoft and IDC. At IDC, Michelle reimagined marketing as a strategic growth engine, launching a bold new brand identity, spearheading the company's first GenAI initiatives, and aligning brand, demand, and strategy to drive global impact. Michelle is a frequent speaker on marketing leadership, AI, and purpose-led growth. A lifelong learner, she's also a runner, rescue dog mom, and dark roast devotee. Her best ideas rarely arrive in a meeting, but often hit mid-stride or mid-walk. Michelle Boockoff-Bajdek on LinkedIn: https://www.linkedin.com/in/michellebb/ ---------- Resources ---------- Sitecore: sitecore.com The Agile Brand podcast is brought to you by TEKsystems. Learn more here: https://aglbrnd.co/r/2868abd8085a9703 We're proud to be a media partner for #MAICON26 - Oct. 13-15! Learn how AI can power your marketing and business and help you grow smarter. Use code AGILE150 to save! https://aglbrnd.co/r/7fe458ced0f04658Reach your customers with Reddit. Spend $500 in ad spend, get $500 back in ad credit! Learn more: https://advertalize.com/r/491818c79fb1873fDon't miss We Make Future - the International Festival of Innovation in AI, Tech, and Digital Marketing, June 24-26 in Bologna. Learn more: https://aglbrnd.co/r/c80991afff416bb2The most influential minds in software, AI, and engineering leadership will be at WeAreDevelopers World Congress North America, September 23-25 in San Jose. Learn more: https://aglbrnd.co/r/60a7299222a7bcf1 Enjoyed the show? Tell us more at and give us a rating so others can find the show at: https://aglbrnd.co/r/faaed112fc9887f3 Connect with Greg on LinkedIn: https://www.linkedin.com/in/gregkihlstromDon't miss a thing: get the latest episodes, sign up for our newsletter and more: https://aglbrnd.co/r/35ded3ccfb6716ba Check out The Agile Brand Guide website with articles, insights, and Martechipedia, the wiki for marketing technology: https://www.agilebrandguide.com Hosted on Acast. See acast.com/privacy for more information.

AI is reshaping work by expanding roles, increasing multitasking, and accelerating productivity—but companies must set healthy guardrails and rebuild entry-level pathways as AI automates traditional starter tasks. That's the key take-away message of this episode of the Wise Decision Maker Show, which describes how generative AI is reshaping work exactly as expected.This article forms the basis for this episode: https://disasteravoidanceexperts.com/generative-ai-is-reshaping-work-exactly-as-expected/

We're announcing AIEWF speakers this week! Take the AI Engineering Survey!Today's guest Ethan first joined us for the LS Paper Club as the lead on NVIDIA Cosmos World Model, but then joined xAI and built Grok Imagine in 3 months:He comes back on Latent Space with some nuclear hot takes: that Video Models primarily get their intelligence from LLMs, not from training on video data, and that the next frontier for truly interactive, realtime, long-horizon world models is to work on LLMs (perhaps Interaction Models as well…)Put it this way: In the near term, the next Sora won't be a better video model, but a video agent.Generative Media may more closely follow the evolution of AI coding which went from focusing on one-shot output performance and cost, to multiturn reasoning and planning models for agents and systems that can plan, edit, test, debug, and submit PRs.At a certain point, coding models got so good that the only significant next step to improve performance was handling the orchestration of these models.Now as the performance of video models increases significantly across realism, consistency, & prompt adherence while becoming more cost efficient, the next evolution of video generation may also be systems that can plan, generate, edit, critique, and iterate across an entire creative task. In this episode, Ethan joins swyx and Vibhu to unpack what it actually takes to build frontier image and video systems: data, VAEs, diffusion transformers, audio-video alignment, inference speedups, and the hidden cost of storing and moving massive video datasets. From building NVIDIA's Cosmos world model to joining xAI as Grok Imagine was being built from zero to one, Ethan He has been at the center of some of the most important work in video generation, multimodal models, and real-time world models.We go deep on Grok Imagine, how a small xAI team shipped its first multimodal video model in three months, why iteration speed matters more than almost anything in model development, and why many of the biggest gains come from fixing tiny bugs in data and training pipelines. Flipbook: The future of VideomaxxingVideo agents are almost a sure bet to be the trend in the coming year. We end with a glance at what's beyond video agents:Flipbook caused a minor sensation this year when it was released, but most treat it as a fun demo. Ethan takes it very seriously — with the speed and cost of inference coming down every year, the future of custom video JIT UI is closer than you think. We talked about why videogen models may become the front end of AI, how generative UI could replace traditional HTML/CSS, why world models need to be real-time, interactive, and long-horizon, and why the future of video generation may depend more on language models and agents than on diffusion alone.We discuss:* Why fast iteration mattered more than meetings* Why small training bugs can drive huge model quality gains* Why coding models may make compute the bottleneck again* How image and video models are trained with synthetic captions* The role of VAEs and latent space in frontier video models* Why image models are the foundation for video models* The tradeoff between temporal compression and real-time interactivity* Flipbook, Neural OS, and the future of generative UI* Why future interfaces may go from user intent to pixels* The hidden cost of training video models: storage, egress, and GPU hours* How step distillation and consistency models (like OpenAI sCM) makes video inference orders of magnitude faster* Grok Imagine 0.9 and large-scale audio-video generation* Why audio-video alignment is harder than text-video alignment* Ethan's definition of world models* Reference-to-video, video extension, and long-context video generation* Why xAI's research communication undersells Grok Imagine* How xAI culture shaped the speed of development* AI watermarking, SynthID, and detecting generated media* Why prompt rewriting matters for video models* Grok Imagine Agent and the rise of video agents* Why language models may unlock better video generation* Robotics, physical AI, and embodied world models* Why Ethan left xAI and shifted focus toward LLMs* Self-managed context, memory, and the next frontier for language modelsEthan He* LinkedIn: https://www.linkedin.com/in/ethanhe42* X: https://x.com/EthanHe_42Timestamps00:00:00 Introduction00:01:25 From NVIDIA Cosmos to xAI00:03:24 Building Grok Imagine from Zero to One00:10:07 How Image and Video Models Are Trained00:18:53 Video Compression, VAEs, and Real-Time Tradeoffs00:22:10 Generative UI, Flipbook, and Neural OS00:32:10 The Cost of Training Large Video Models00:37:04 Distillation, GANs, and Fast Video Inference00:41:21 Audio-Video Generation and Grok Imagine 0.900:48:34 What Makes a World Model?00:55:51 Reference Videos, Long Context, and Video Memory01:00:11 xAI Culture, Research, and First-Principles Building01:09:45 AI Safety, Watermarking, and Prompt Rewriting01:13:10 Video Agents and AI-Assisted Creation01:27:32 Why Language Models Unlock Better Video01:31:15 Robotics, Physical AI, and Embodied World Models01:32:38 Why Ethan Left xAI01:34:16 Self-Managed Context and the Future of LLMs01:38:43 Ethan's Career Path and Closing ThoughtsTranscriptIntroduction: Ethan He, Latent Space, and the Path to xAISwyx [00:00:00]: We're here in the studio with Ethan He, most recently of xAI. Welcome.Ethan [00:00:10]: Thank you. Glad being here.Swyx [00:00:11]: We're also here with Vibhu. you were first coming to us or joining the latent space world because you were working on Kosmos at NVIDIA, and you did a paper. We loved it. you presented it as well, so thank you for doing that.Ethan [00:00:23]: I've actually, I also presented the MoEs twice at latent space.Swyx [00:00:29]: How did you actually hear about us? Did we reach out to you? Is that how it worked?Ethan [00:00:33]: No, actually, I-- the community. Like I realized, oh, there is this online community that people talk about AI and also learn from each other through papers every week through the Paperclip. It's very nice.Ethan [00:00:49]: I learned a lot.Swyx [00:00:49]: I think three years stop. We haven't stopped even on Christmas and New Years. many weeks I want to stop but it keeps going.Vibhu [00:00:58]: No, that was good. I think you had posted that you worked on a paper, and I was “Oh, very cool. We have Paperclip. Present then.”Vibhu [00:01:04]: But I might have reached out to you after.Swyx [00:01:05]: you-- because it's an amateur club, right?Swyx [00:01:08]: so it's very unusual and but we have sometimes paper authors come by and actually explain the paper. Today we just did, the poolside paper, which was apparently very good.Vibhu [00:01:18]: Came out yesterday.Vibhu [00:01:19]: pretty interesting, right? Fully open. They talk about everything, systems. So it's a good one. We'll, we'll recommend people to read it.Swyx [00:01:25]: Bring us up to speed on your transition to xAI, ‘cause I actually don't even know when you joined. just like tell the, tell the story about the sort of transition.From NVIDIA Cosmos to xAI: Scaling Video and World ModelsEthan [00:01:34]: Before xAI, I was working on Kosmos world model as in-- at NVIDIA. So Kosmos is, it's a giant video foundation models that can-- that aims to simulate the world and for-- it serves as a foundation of-- for all of the roboticists to build on top of. There, once I built the Kosmos one, I realized as this thing also has a scaling law similar to language model, we need to scale up the video models further. that's, that's why I realized I need to move to somewhere with much more compute resources. That's how ISwyx [00:02:13]: Than NVIDIA?Vibhu [00:02:14]: The GPU rich came themselves.Vibhu [00:02:19]: And timeline-wise, when was Kosmo? It was pretty early, right? It was open world model, open paper, everything.Ethan [00:02:25]: It was end of twenty-four.Vibhu [00:02:28]: End of twenty-four.Ethan [00:02:30]: Then at mid twenty-five, I moved to xAI. At that time-- I joined about the time when xAI was about to build video models and in multi-model models. There were no infra, no data, and no model, and it just-- as a few engineers, we built it in three months and released the first model, Grok Imagine zero point nine.Ethan [00:02:55]: And since then, I keep working on video models and move more from training and to post-training of the video models. For example, like a reference to videos, kind of like the cameo feature and, video extensions. And, before I left, I worked on a world model, leading a small team to focus on the real-time long horizon video generation.Building Grok Imagine From Scratch in Three MonthsSwyx [00:03:24]: Can you give like a rough roadmap of okay, you're on a brand-new team. Grok previously was only text, or they partnered with BFL for their image gen stuff. What do you-- what are the building blocks, right? You have compute, data you can procure somewhere. Like just what are like the sequence of things that people should think about when you're setting up a new team?Vibhu [00:03:43]: actually even deeper, not just data you can procure. You guys had to go through getting the data too, right? So you shipped it pretty fast, but yeahSwyx [00:03:51]: three months is likeVibhu [00:03:52]: From everythingSwyx [00:03:52]: actually like very surprisingly fast.Ethan [00:03:55]: One thing I say like thanks to my experience at NVIDIA, ‘cause first time when we were building Kosmos together, we built it, for about a year. So this is like the second time I do it. Roughly have an idea, what to do. I say the most important thing is the talent. Everyone were very strong and clever, very close with each other towards a common goal. So that speed up things a lot. So you reduce the communication bandwidth among people, and everyone can work towards the same goal. It's, it's like every day there's not that much meetings on the calendar, like maybe like a, like a sync a day, and after that it's, it's just all building. It was pretty fun at that time.Ethan [00:04:47]: And another thing is that xAI has very strong foundations of like data inference, model inference, and the supporting there can help the model develop a lot. When I look at, training models, I don't so actually the top important thing is like how many, how many iterations can you do, per day? and the more iteration can you do, you can, you can train the model much faster. So if you have very strong infra and you have a lot of compute, you can, you can train these models in very short period of time. That can give you a much larger buffer to, for errors, and it also gives you the opportunity to spot more bugs.Iteration Speed, Compute, and Debugging Model PipelinesSwyx [00:05:46]: What is an iteration? Is it like a few hundred steps or what are youEthan [00:05:50]: Let's say just the train-training the model, like from acquire new data and maybe design new algorithms and train a new model, maybe at smaller scale orSwyx [00:06:01]: So cycle time for like any hyperparam that you're searching.Ethan [00:06:04]: Cycle time and tune to like eval this model. Is this model better than my previous iteration?Ethan [00:06:11]: SoSwyx [00:06:11]: So it's like before you, someone had already set this up that you can iterate very quickly.Ethan [00:06:15]: I think the foundation there is extremely good forDeveloping and research models.Ethan [00:06:23]: And often I find is it-- this is kind of boring, but like a lot of the improvements does not come from new algorithms. It comes from finding small bugs here and there in the data pipeline, in the, in the model training pipeline. Those give, those give the biggest boost to the model quality.Vibhu [00:06:46]: It's interesting, right? So you say it's like small team, less communication bandwidth, but also a lot of quality is like find little bugs. It seems counterintuitive, right? You have a lot of people, you can iron out more of those, but it's interesting to see the other side, right?Swyx [00:07:00]: I also wonder, have you-- do you try using LLMs to look for bugs? I don't know.Ethan [00:07:05]: I remember at that time it was mid two thousand and twenty-five, so it's the coding model wasn't quite there yet. I remem- I remember like December two thousand and twenty-five, it was extremely good. Yeah, I've been, I've been using it at that time. It's, it's helpful. sometimes it produce codes that are kind of difficult to maintain, even though like the first time it built something extremely fast. But it gave the, like a spaghetti code, thousands of lines that I couldn't maintain, and the LLM itself couldn't figure out what's, what's wrong and how to improve on top of it. But now I find it much better. Yeah, I want to bring up another point here is now coding models are much more efficient and can help us implement stuff much faster. Compute might become a bottleneck again because previously, like if you want to train a new model, say you want to generate new synthetic data and then or write a new algorithm, it might take a few weeks. And during that period of time, you don't-- you might not have experiments to run. But now you can build that thing within a few hours, then you can immediately train a model.Ethan [00:08:24]: Now you have to have enough compute to try all of the ideas. So compute might be the bottleneck of iterating speed again.Swyx [00:08:36]: yeah, I actually, honestly, I think it's like kind of a stressful job because you're “Well, I should be trying everything, and if I'm not, then I'm not doing my job well.”Vibhu [00:08:48]: there's also the stress of you're eating thousands of GPUs per hour, which is very expensive and, compute can go to other researchers.Swyx [00:08:56]: You got the daddy Elon toVibhu [00:08:57]: You got daddy Elon.Ethan [00:08:59]: It wasVibhu [00:09:00]: But there's still finite amount of compute, like you want to use it, you want to use it well, you want more of it.Ethan [00:09:06]: That was quite stressful indeed. Yeah, I think one thing is the-- with coding models now, like a lot of these jobs can be automated, which is much better. A second, it's a, it's a marathon, so you got to maintain good health and, a regular schedule.Vibhu [00:09:28]: It's, it's hard to hear that when you shift from zero to nothing in two months.Swyx [00:09:32]: and, I think obviously the culture at xAI is very famously, people work very hard. one thing I did want to dive into, in our-- in the notes that you, that you sent ahead of time, you had specific comments about the cost of Video Gen training. presumably this is on the Colossus-1, right? the two hundred megawatt cluster. Any whatever you want to just share on that.Vibhu [00:09:54]: I think there's, there's three things we're talking about, right? So there's Video Gen, there's also the Image Gen model that you put out. Do you want to like complete the, okay, so zero to one, you have a few months. Just what are the stages of create Image Gen model?Swyx [00:10:06]: Oh, yeah, maybe I got distracted.How Image and Video Models Are Trained: Synthetic Captions, Tokenizers, and VAEsVibhu [00:10:07]: Sorry. and then, from there's Video Gen, there's Audio Gen. Would love to get into those next. But what is that first few months like? So small team, a lot of bugs, iterations, but what does it look like? Do we take something off the shelf? Do we just get data compute? What's, what's the few months like? How do you go to state-art Image Gen model? How do you just start?Ethan [00:10:28]: I cannot comment specifically how xAI did, but it's, it's a quite standard process. I can draw some, examples from Cosmos. So mainly it's building a video model, you actually need to build a image model first. And building these two models, the data you need is a hundred percent synthetic pair of language and image or language to video. Because on the, on the internet, actually, the videos don't naturally associate with text. So you can say, oh, like on YouTube, you have the title and you have the description and the commentsSwyx [00:11:11]: TitleEthan [00:11:11]: of a video, but usually they're not relevant to the video itself. And say maybe like the video is a natural scene of mountains or something, and the title is, I'm so happy today.Ethan [00:11:26]: So they have they have no correlation at all. So the first step is to, you have to generate synthetic pair of language with the videos. So you gather videos from the internet, and you use a VLM to caption the videos. So that part, here's a question, like how do you, how do you gather VLM to begin with? So if there's noSwyx [00:11:55]: You, so you fuse the model, right? LikeEthan [00:11:57]: Say if there's no like VLM exists, like how do you generate the text to the beginning, right? It's, it's impossible.Swyx [00:12:04]: I see.Ethan [00:12:05]: In the beginning, it's like you ask human to describe the video as detailed as possible.For example, you ask them to describe everything, like all objects, all characters, and all interaction and dialogues in the, in the videos. So that's in the protocol of Cosmos labeling. We require the objective we give to the labelers was that you have to describe the video as detailed as possible, such that a blind person hears a blob of text can reconstruct what the video is like from their head.Swyx [00:12:43]: Video or image? You're talking about images.Ethan [00:12:44]: Video or image, either one of them.Vibhu [00:12:47]: This was pretty common when we went from clip and DALL-E, right?Vibhu [00:12:51]: It's all training on really detailed captioning of images. So same is applied to video, but insteadEthan [00:12:57]: same appliedVibhu [00:12:57]: of using multimodal model to pass in video images and write rich descriptions, you can alsoSwyx [00:13:04]: I think there's this traditional perspective of supervised, or, very highly human curated thing. I feel like there's a unlock with unsupervised, right? Where like you have enough to bootstrap that you can just throw common corpus on it or, whatever. like unsupervised vision and language pairing, right? Like where you just have, interspersed image and text and it just learns. To me, that is the VLM breakthrough that is different from the clip, different from the LM era.Ethan [00:13:36]: It's interesting to see that you kind of need both data.Ethan [00:13:41]: For example, for theSwyx [00:13:41]: You need it to bootstrap it up. YeahEthan [00:13:43]: for the generative model training, there's also usually like a small percentage of unlabeled data. So the model is instructed to generate a video without any text instruction. That can also help the model generalize. So after this stage of generative synthetic pair, so, one important common step is to train a compressor or a tokenizer of the image or videos. So because, if you train-- If you can technically, theoretically train image or video models on pure pixels, but the problem is that the, it's, it's a lot of tokens. So like one image, it's, a thousand by a thousand, it's like one million tokens, one million pixels. It's impossible to train transformer on that. So it's, you need to train a tokenizer, which can go from image to latent space and latent space back to image.Swyx [00:14:45]: That's why we named the podcast.Swyx [00:14:48]: But, basically, you're talking about vocabulary science.Ethan [00:14:50]: so vocab.Swyx [00:14:51]: And so, what is, what is imp-- like a million is impossible?Ethan [00:14:54]: In generative models, the vocab is continuous. It's a continuous space. We can think about like you map an image to a vector. It's a, it's a fixed length vector. It's sixteen or forty-eight, something like that. And then you map that vector back to the image space. And the mapping is, has-- The mapping is patch-based. So you say you haveEthan [00:15:22]: a sixteen by sixteen patch and you match, you map that patch of pixels into this latent space.Swyx [00:15:29]: We've covered thisVibhu [00:15:30]: This is like the vision transformersSwyx [00:15:32]: VAEs,Ethan [00:15:33]: VAEs.Vibhu [00:15:34]: You basically compress your input, you do your generation, you're reasoning all that generation in smaller dimension, and then you project back out.Swyx [00:15:43]: VAE is a form compression, but I think the for me, the patching thing is from VIT, right?Ethan [00:15:48]: You can make those.Swyx [00:15:49]: Literally the, yeah, the paper is titled like sixteen by sixteen is all you need. something like that. and then I think also, people make a lot of comparisons with this kind of patching with convolutions.Swyx [00:16:02]: Which is you're, you're kind of re- reconstructing the old paradigm with the new.Ethan [00:16:05]: Actually, in VAEs, there are, there are both convolution networks and transformers. You can actually do both.Ethan [00:16:14]: After this VAE, so what you've got is you've got latent space tokens and you've got the language tokens. So now the training of the diffusion transformer, usually generative models use diffusion transformers. It is actually quite standard. It's, it's very similar to how you train a language transformer models. It's not that much difference. It's just the tokens, the visual tokens in, visual tokens out. The only difference is there's a denoising process. So you train the model to unmask some of the noise. So you add, you add random noise to the visual tokens, and then you train the model to remove those noise to generate the clean tokens. Any inference, the model can iteratively remove noise from a hundred percent noise.Swyx [00:17:12]: And then there's also, to speed things along on the tech tree of diffusion, there's CFG, and then there's, there's also, latent diffusion that, there's, there's someone in there. I think, somewhere along the line, obviously, like stability and all these other guys, pioneered a lot of this, architecture. I don't know if you want to get into that or just, or do the video side up to you.Bootstrapping Video from Image Models and Temporal CompressionEthan [00:17:37]: After you train such model, such image model, the reason it's a, it's a foundation for video models is that image models are cheaper to train, and they have much denser connection between language and text. So, sorry, language and images. For example, you train a billion, you train on a billion images, and there's a mapping from the text to the image. And the cost to train the same, like the, a billion, a billion text to a billion videos, that's much more expensive because videosNaturally have more tokens than images. Because the diffusion models, their understanding of, language purely come from this mapping. So if you don't have enough mapping, so if you only train on like a ten million videos or something, there-- you might not see enough language tokens in your training, so your model does not understand human intention enough. So that's why you really-- you train-- you first train this image diffusion models, and then you bootstrap the video model from there.Swyx [00:18:53]: One thing I did want to ask, because I-- actually, I think you're, you're the first per-- video model person I've ever talked to, I think. we've, we've like talked to Luma and all those folks. There's all these tricks in video compression where basically frame by frame there's not that much difference, so actually you don't have to regenerate or save the whole frame, right? but I think MP4 compression or something else like that.Swyx [00:19:16]: is it tempting to use that? Or as far as I can tell, everyone just treats it as, “No, we would just generate every frame.” Is that roughly the state-art?Ethan [00:19:27]: There are a few different approaches. Let's say first, like you want to just directly use MP4 compression and use that as the tokens for the transformers to train, right? So people actually have tried that, but the main challenge is the latent space for the MP4 tokens were not, were not very comprehensible for the models. It's, it's extremely hard to train on that. And there's aEthan [00:20:01]: So that's why they created VAEs, which creates more continuous, latent space, so the models can understand that latent space and learn from it much easier. Even within the VAEs, there are different difficulties of the latent space. So you can imagine something the simplest, the most naive VAE is like you have an image, and you just shuffle all of the images into a, into a vector. So you don't need to train any VAEs, right? But that latent space is extremely hard for models to train on top of. That's why there are some debate on like how do you compress the tokens. So you mentioned like you can compress frame by frame. Also, you can compress, the temporal dimension.Ethan [00:20:52]: The difference is if you compress the temporal dimension, you get a much higher compression rate. Because there's temporal redundancy between frames, because, this frame and the last frame, likely they are mostly similar, so there's only some small difference. for example, I think in 12.1 VAE, they have like a eight by eight by four compression rate. So the four temporal tokens are compressed into one tokens. That can save a lot of, save a lot of the context length. If you do it frame by frame, you have to do maybe like eight by eight by one. Your context length will be four times larger. That being said, the benefit of the frame-- per frame compression, we might come back to this later, is, real-timeness and interactivity. ‘Cause if you, if you strain the output of the model, frame by frame, you can-- the model can respond to any user request immediately. So if you have like a temporal four compression, four times compression, thenSwyx [00:22:06]: It might be laggyEthan [00:22:07]: there's a lag there in nature.Swyx [00:22:10]: So you're very pilled on this. let's just go ahead and bring it up ‘cause we have the visual prepared anyway. There's some frontier applications of real-time video gen. So Flipbook is one of the examples that went viral recently, right? What is Flipbook?Real-Time Generative UI: Flipbook, Neural OS, and Diffusion Front EndsEthan [00:22:23]: Flipbook is kind of like a web brow- web browser. You can see like it has the web bro- browser UI on top. The difference is all of the UIs are generated by generative image model in real time, and anything here are fake. But you can, you can explore inside this wor- this imaginary world. Say like we-- here we have engineering the Great Pyramid. Like the model generates this for us to understand how it works, and if we want to navigate around and understand further, we can click on some of the, some of the description here, and the model will generate a new page, new subpage describing the details we want to know about.Swyx [00:23:14]: So it's basically kind of we're playing a video, but it's pausing for our next interaction, and then it just plays the next thing based on our interaction.Swyx [00:23:23]: Which is kind of cool.Vibhu [00:23:25]: and you kind of decide your story. So this was, how do you make a pyramid? levering technique seemed interesting, right? It shows how do you take Okay, I want to know what is thisSwyx [00:23:35]: The demo, the demo tweet had more animation between frames.Vibhu [00:23:38]: I think it's just skipping,Swyx [00:23:39]: Oh, it's just skipping a lot of frames.Ethan [00:23:40]: they also have a video modeVibhu [00:23:42]: It takes a lot. There's a lot of peopleEthan [00:23:42]: but, a lot of people are using it.Ethan [00:23:45]: So it's not available.Vibhu [00:23:46]: There's a live video stream. We can try,Swyx [00:23:50]: So this is an example of the kind of future that you see at the extreme. We don't-- we're obviously not in it today.Swyx [00:23:56]: But in a world where inference is completely free this is better than generating code and text?Ethan [00:24:02]: So this is, this is a final state of where Viva will be at for word model, I think. Imagine internet doesn't exist, and then you type in google.com. Like what should, what should, what should a model show you?the model can imagine something, and this is what the model imagine. And these web pages, they completely do not exist. So I think as the inference costs come down, we are going to have generative UI for everything. If you think about how the coding model works, so they write code for a web page, and they render the code might be con- converted into binary, and the binary render the pixels on the screen. So we in machine learning, every time we have some breakthrough, obviously it's, it's more intuit. So why don't we have like user instruction to the pixel directly? So the generative UI will be user intention to the pixels directly. And say like even if I want email, let's say everyone have the same interface, but I want, I want it slightly different. I want the email to show to me like a TikTok, so I can swipe left and right for the emails. And or maybe you want something else. We can have completely different things. Or like I have I'm looking at, Instagram stories, and I don't like the Like button. I always may click it. And, generative UI resolved it. So it's going to be a revolutionary replacement of the interface. So in the future, we might have much more powerfulEthan [00:25:50]: LLMs and coding models running behind the scene. And in the, in the front-end, the diffusion model will actually be the front-end to show stuff to you. That's how I imagine it.Swyx [00:26:02]: Diffusion front-end, deterministic back-end.Swyx [00:26:04]: Something like that. I find that very expensive, but,Vibhu [00:26:08]: I find it interesting you called LLMs writing code on the back end deterministic, but okay.Swyx [00:26:14]: you write it onceVibhu [00:26:15]: Compare it toSwyx [00:26:16]: And then you execute.Ethan [00:26:17]: If you think about the cost, say, let's say H100 costs $1 per hour, and if you use this eight hours a day and thirty days, so, every month you're paying this two forty, you'll actually not wanna pay for that. That's even more expensive than Cloud Code Max. But if you think about the compute costs come down like two times every year, and I think the future will likely arrive like within few years.Vibhu [00:26:49]: It's everything, right? compute cost comes down, compute gets faster, model gets smarterEthan [00:26:54]: More efficientVibhu [00:26:54]: model gets smaller.Swyx [00:26:55]: I don't know why you say two times, ‘cause I think it's like 100 times. In language models, it is roughly one hundred to a thousand times every twelve to eighteen months, for the same given level of LMSys, ELO.Vibhu [00:27:08]: That's a net of everything, right? That's model performance alongside compute. So different than just compute costs come down. But, a very interesting future.Swyx [00:27:19]: So the web designers will have to shout out that accessibility is an issue, right? how do you deal with screen readers or whatever. But yes, this is higher bandwidth storytelling than anything you can possibly generate with code, right? So I think that's the rough idea.Ethan [00:27:34]: And I'd like to add a little bit that so human naturally have the maximum bandwidth when we are looking at things, look at videos, and we also have maximum output bandwidth when we are talking. So in the future, it might be something like we talk to AI models, and the AI model responds back with a generative UI. So that would be the maximum input and output bandwidth to interact with AI models before neural link happens.Vibhu [00:28:06]: And it's also very custom, right? Some people are very visual, some people are not as visual, right? They prefer the text. But the best thing about generative UI, right, it can also be text.Swyx [00:28:17]: There's another project that we wanted to highlight, which is the Neural OS. Kinda similar idea, but here you're literally operating, simulating an operating system with a video model.Swyx [00:28:27]: and you can play Doom, you can do Firefox. I find this like mildly less impressive, obviously, because it's an OS that I can run.Swyx [00:28:37]: But here everything is imagined.Vibhu [00:28:40]: I was, used to the Command+W to close the Firefox tab. It didn't crash. That's why I saidSwyx [00:28:45]: It's too immersive.Vibhu [00:28:46]: It's, it's too immersive for me.Swyx [00:28:47]: Too immersive.Vibhu [00:28:48]: I wanted to close the tab.Vibhu [00:28:49]: But yes, I can play generated diffusion.Swyx [00:28:51]: this is shockingly fast.Swyx [00:28:54]: Because I remember there was a demo about like maybe one to two years ago. Someone tried to do the first-person shooter with a image model. There was no consistency. It was very slow. But here it looks like realistically it's-- this is Doom.Vibhu [00:29:07]: I think there's two sides to that, right? There's okay, what is running a game? The heavy part of it is actually the game engine, all the lighting, all that stuff, the graphics. This is just kind of video, right? Like we've solved consistency. This is still, it looks like a few years old image generation. There's some temporal consistency, but it's, it's kind of just images stitched together as frame video. But it's a good visual representation to pi- to picture the future you wanna see, right? that's, that's what I see in these more so.Ethan [00:29:38]: This reminds me of how the video models gets better and better. So Neural OS is kinda if you just look at it feels like it's just a crappy version of the, like the Windows we could have, right? And, but the difference is, so the model, this model is overfitted on the existing operating systems. It can generate nothing different than that. But it's actually also similar to video models. So when we are training these video model, image model, we train them on internet. There's no imaginary supernatural stuff on the internet. But once we train this model, you can prompt the model to generate something supernatural that have never existed in the data set. So if you train your Neural OS or neural computer on the standard screen recordings on the entire internet. The model can imagine completely new interface to interact with the computer.Swyx [00:30:43]: This is one of those things that is magical to me. usually generalizing out of distribution is bad, but somehow we have learned some kind of internal world model that you say, this plus, but it looks like rainbows and butterflies, it'll do it and it will kind of make sense.Swyx [00:31:03]: So yeah, that's kind of cool. Yeah, I don't know if there's any comment more on there. I do, I do wanted to, I did wanted to touch a little bit more on the model architecture stuff, which I think you were getting. It's, really fascinating. We don't get a chance to talk about this enough. So one of the papers that we covered, we've covered every annual, segment anything release. and I don't know if you follow-- you're a computer vision guy, so youEthan [00:31:26]: I knowSwyx [00:31:27]: . So they did memory attention, which is kind of interesting. And I always think, anything where you can, across the temporal dimension, keep some consistency, I think it's, very fascinating, and I don't know if Basically, does that-- the CV side bleeding into video gen side, I think is underexplored, right? we talk about it for labeling, but actually you can borrow the architecture itself.Ethan [00:31:50]: There's, there's also complete different approaches, right? you brought up the term world model, so we went from video model to world model. There is diffusion, but there's also other approaches that people are doing. So maybe we get into those after as well,?Swyx [00:32:03]: He has a whole definition of world models and stuff. I feel like we threw a lot at you. Whatever you want to comment on.Why Video Models Are Expensive: Storage, I/O, and Training ScaleEthan [00:32:10]: I think one thing that we should actually comment back on is okay, so we were talking about the steps to train image gen to video model. One thing we don't see as much of is okay, you brought up the delta in training data, right? SoEthan [00:32:24]: you won't have as much a video model might not generalize, but what is the cost of training a large video model? So we know for LLMs roughly, okay, even like the poolside thing that came out today, right? It's a Gemma level model trained on roughly forty trillion tokens at this many H200s over this much time, right? You can see what is the exact cost of that. So how many GPU hours over how much H200 costs? So how do we do the back-end math of, same thing for video models, image models. How do you, how do you kind of break that down? I can share some back-envelope calculation. So surprisingly, video models is-- the cost is very-- is comparable to language models and obviously the largest scale is language model, maybe like a medium scale to language models. I said just storing the videos alone, it costs a lot. You can, you can maybe look up on AWS or something.Ethan [00:33:20]: You really, say if you have a billion videos and let's say, let's just say like each video, like five megabyte, then you need five petabyte to just store those videos. And also remember we talk about you use a VAE to compress the videos, and you also need to store, typically you need to store those continuous feature, in-- also in your storage. That's also comparable size with the videos themselves. So just storing these videos and the features is tens of petabytes alone. And,Swyx [00:33:58]: I just, I just looked up the calculation. Five petabytes on S3 Standard is one hundred K per month.Ethan [00:34:05]: AndSwyx [00:34:05]: It's comparableEthan [00:34:05]: and you needSwyx [00:34:06]: AndEthan [00:34:06]: And then like tens of petabytes, two hundred K. And even more expensive is you have the ingress and egress.Swyx [00:34:13]: Oh, yeah.Ethan [00:34:14]: Like you-- through the internet. You have to just to download those videos, I believe it's, it's more expensive on AWS than just storing those videos.Swyx [00:34:25]: Storing, yeah.Ethan [00:34:25]: And each training runs, you probably need to pull them once. If you train multiple times, it's, it's even more than that. So it's like just storing the network, those costs is just, it would be a few, a few millions per month to just storing everything, not to mention the GPU cost.Ethan [00:34:45]: AndSwyx [00:34:45]: my side tangent, the compute rental, like GPU rental is very efficient. There's one side, okay, you can be XAI and build your data center. Should we not just build our, storage compute as well? LikeEthan [00:34:57]: Of courseSwyx [00:34:57]: cloud cost compared to just,Ethan [00:34:59]: You save so muchSwyx [00:35:00]: store. Yeah, exactly.Swyx [00:35:01]: Especially with like egress and stuff. So.Ethan [00:35:04]: That's a good idea, but it also comes to-- there are some of its own challenges.Swyx [00:35:09]: Of course, of course.Ethan [00:35:10]: like people who build the GPU data centers, they might not expect this much, storage. And yeah, people build storage, typically they just build it somewhere with just CPUs.Swyx [00:35:23]: I just looked it up. Five-- AWS only charges for egress, not ingress. Tier five for five petabytes is two hundred and thirty K.Ethan [00:35:32]: Even more expensive than the storage.Swyx [00:35:34]: But storing is per month, right? You check in, then you cannot check out. so it's so cool. It's okay. So there's that side.Ethan [00:35:41]: So the TLDR, my backhand mathSwyx [00:35:42]: Data is larger than you think. Yes.Ethan [00:35:44]: my backhand math of GPU hours times GPU cost is also very much, I'm missing some storage.Swyx [00:35:49]: You're also-- you're basically like also more IO bound than normal training.Swyx [00:35:55]: Yes. ‘Cause like data loading, so caching everything, it becomes super important.Ethan [00:36:00]: So in Cosmos, we did a lot of optimizations to make it not IO bound. So, speaking of the training, actually training the model, the GPU cost, if you look up like the open source model, how big these video models are, I think like LTX has nineteen B parameters. That's a dense model. And people are also exploring, MoEs, so it might be twenty B active and, like a hun- hundreds B, total. So that's, that's even-- that's similar size as medium-sized LLM models. And if you, if you look at number of tokens-Uh, we disclose that in Cosmos. It's also like tens of trillions of tokens on the visual tokens. So putting this together, the cost of, training these video models, it's actually comparable with LLMs. Not to mention, the infra is slightly different from LLM, so it might be less efficient to train these models.Inference Speedups: Step Distillation, Consistency Models, and GANsSwyx [00:37:04]: Do you get the benefits of traditional diffusion speed-up? So for, images, there's LCM, LoRAs for, fine-tuning. There's, there's a lot of stuff that's beenEthan [00:37:15]: Flow matching.Swyx [00:37:16]: there's flow matching. There's a lot of stuff that's been done. there's some overlap that applies to diffusion on the inference side and stuff or?Ethan [00:37:23]: so the difference-- the inference side is a completely different story.Ethan [00:37:28]: I think for the training side, it might be a little bit hard to reduce that cost. And for the inference side, the biggest gain is from the distillation of these models. You can-- It's called step distillation, slightly different from knowledge distillation in LLMs. So you-- Typically, for flow matching models, you need like 100 steps or something. Like a distortion model even need even more, like 1,000 steps to generate a good image or video. A step distillation is try to learn to generate fewer step from the model itself. It's kind of like now we-- you use the full model to generate in 100 steps, and then you take a model that only generate 10 steps and let that model to learn from the perfect one.Ethan [00:38:25]: why this workSwyx [00:38:27]: Strong to weak seemingly.Ethan [00:38:28]: It is. It's kind ofSwyx [00:38:29]: DistillationEthan [00:38:29]: kind of like strong to weak. the-- from the modeling perspective, the strong model, the teacher model is trying to model the image and videos of inter-internet, and that distribution is extremely complex. But the step distilled model is just trying to learn from the teacher. The teacher is a model, and the size is fixed, as the distribution is much simpler than the whole internet. That's the intuition I have why step distillation can work. So usually these models serve in productions, they only run in a few steps. In Cosmos, I believe we have, we have like four step and eight steps. If you do some simpler task, image-image translation, it can even run in fewer step, like one step in Cosmos Transfer.Swyx [00:39:22]: I think this is the same intuition that guides a lot of the consistency model work. I sent you a link for, SCM. I don't know if you covered that. To me, that was actually one of, the most impressive papers I've ever seen from OpenAI.Swyx [00:39:34]: That this is the unifying grand concept of consistency models. I don't know if you have any comments on this.Ethan [00:39:41]: So there are, there are a few different approaches,Swyx [00:39:46]: Oh, yeah. Here it is.Swyx [00:39:47]: Two steps versus twenty or 100 steps, whatever. It's already done.Ethan [00:39:52]: So there are, there are a few different approaches, for example, consistency model, and there are also Actually, we shouldn't forget GAN. So GAN, actually, that was, that was the OG ofSwyx [00:40:05]: OGEthan [00:40:05]: step distillation ‘cause it trained just one step to begin with. So actually, a lot of, uh-- For example, there's a distribution matching distillation which use, which uses GAN, as one of the laws for distillation. It-- GAN just tells you, “Hey, generate an image,” and thenEthan [00:40:31]: it has a discriminator to tell, is this image real or not? So the model, the model just need to learn one of the distribution, not the full distribution. Because in training, the model is asked to reconstruct the ground truth image from the internet, which is extremely hard. And in-- When you're training GAN, it's a step process. It's just a, “Hey, you generate image. Does this image look as real as the image from the internet?” Which is a much simpler task. And, yeah, combining a lot of these approaches together, people typically do that, like consistency model and distribution matching and GAN, and we can get these few step models.Audio-Video Generation and Time AlignmentSwyx [00:41:21]: Then there's one step I wanted to add, which is audio and video.Ethan [00:41:26]: So, Grok Imagine zero point nine, I believe it's, it's a first audio video transmodel deployed at a large scale. SoSwyx [00:41:39]: And that was your first model?Ethan [00:41:40]: that was, Grok Imagine's first model. It's, it's audio video, joint generation. I think the hard part is, the modality alignment, ‘cause before this transmodel, we have, we have text to video alignment. We have this, correspondence between text and video. Typically, most of the VLMs, they understand images and videos. Video's very rare, and they don't understand audio mostly. And if you look at the audio generation on the LLM side, you can talk to them perfectly fine, but if you ask them to sing a song or something, it typically is not very good. Also, they don't have, they don't have music either. The hard part is thatUh, actually audio has two component. It has like a discrete component, a continuous component. The discrete component is like the language.Ethan [00:42:44]: So when we speak, it's just, someSwyx [00:42:47]: It's an ASR issue, yeah.Ethan [00:42:49]: It's, it's text token with some characteristics, I would say.Ethan [00:42:54]: But musicSwyx [00:42:56]: I think the speech guys would disagree with this.Swyx [00:42:57]: Like disfluencies and then,Vibhu [00:43:00]: There's tones you can get angry.Ethan [00:43:01]: Well, I say largely.Ethan [00:43:03]: the mu- but the music is completely different. It's, it's very continuous, and you cannot model them like discrete tokens in language models. this is like the hard part for models is, not to mention we have to align text, video, and audio together.Ethan [00:43:26]: SoVibhu [00:43:26]: How?Ethan [00:43:28]: So significant-- some significant challenges are like-- So first, like we talk about as the VLMs, they cannot understand most of them cannot understand audio.Ethan [00:43:39]: So you have to have some way to do the synthetic data generation for audio. You have to caption the model, and that involve, that involve synthetic data and human data effort a lot. And not just surprisingly, most of the LLMs are very bad at recognizing, like the beat, tone, and the details of the of music. They can, they can give some general prediction of which song is this, but it's very hard to describe the details of the music. like we mentioned in image generation, like you have to describe image as detailed as possible so that someone blind can reconstruct that. So here is like someoneVibhu [00:44:32]: DeafEthan [00:44:32]: someone deaf can reconstruct how the music sounds like without actually listening to it. Maybe you can think of it need to have the-- or they call the script.Vibhu [00:44:49]: Subtitles, yeah.Ethan [00:44:49]: You gotta have all the details of the music, and the dialogue.Vibhu [00:44:55]: So is the challenge there typically stuff like music and audio, or is it just Like is there a baseline? Okay, there's enough data where we can understand, narration, conversation, but there's nuances in audio that's where you hit all the data issues or is it just from stage zero, you just do it all right?Ethan [00:45:15]: So one important thing is like the alignment. So the model, the model has to know like the video and audio, the, uh-- it has to have a time-based alignment, like at which time step the video and the audio token correspond to each other. But we actually don't have this kind of alignment for most of the other modalities. If you think about like text and image, text and video, they are loosely aligned. So you can, you can have a description of what's going on in the video, but you don't have to exactly, You typically don't have exact description, oh, at, time step one second like what happened?Vibhu [00:46:02]: It's veryEthan [00:46:03]: At time step two second what happenedVibhu [00:46:03]: coarse. Yeah.Swyx [00:46:05]: So what was the ideal time step? You have to oblate it, and then it's like four seconds or something.Ethan [00:46:09]: So that comes down to how you design the model to, for the model to be aware of as a time, as a time modality. So the model is like a time aware. And that's something pretty unique if you think about LLMs. So if you ask LLM to complete a task, say they, uh-- you ask them and they will say, “Oh, this task will probably take twelve hours to complete,” and they come back in one hour. Say “I've already spent two days on this and I've exhausted everything.”Ethan [00:46:47]: So the LLMs them-themselves, they don't have a sense of time there.Vibhu [00:46:53]: I actually don't think that's just them not having a sense of time. I think it's somewhat based, right?Vibhu [00:46:58]: Like you tell someone, “Okay, go work on this feature. Go implement this,” there's a general understanding you would have of how long that would take without LLMs working at LLM speed, right? So you think back like two years ago, if I tell you to like build me like a new front end for latent space, have a search bar, have all this, you'll estimate that it'll take a few days, right?Vibhu [00:47:19]: So you tell an LLM, “Go build this.” It'll take me a few days. But I think it's somewhat grounded as opposed to them not having the best-- Not saying that they have a great understanding, but I think that example is like you can see where it comes from, right? You're trained on all over the text.Swyx [00:47:35]: They're, they're trying to estimate what a human would say.Vibhu [00:47:37]: because that's what the, that's what the data kind of represents. It's not themEthan [00:47:41]: It came from the corpus on the internet. People have a estimate of how much time.Vibhu [00:47:45]: And not even just in direct like training samples, right? Just your world understanding of tokens of how long stuff takes, right? Go read a book. It'll take you a while, right?Vibhu [00:47:56]: Even if you do nothing but read a book, it takes a few days. So yeah, LLM, I read it took me a few hours.Vibhu [00:48:01]: It'll take me a few hours to go through this research. But this is a tangent.Swyx [00:48:05]: Somewhat, yeah.Swyx [00:48:06]: This is a train of thought I haven't really expressed until now is, which is basically like a full world model must also be recursive, meaning that the participant in the world model must also be aware that they have a world model. which is like this whole recursive thing down the, down the line. but yes, and that the world model can be wrong and that they need to update it and blah. Yeah. We've, argued this on the, newsletter as well, that there needs to be sort of recursive or adversarial world models.World Models: Real-Time, Long-Horizon, Interactive VideoVibhu [00:48:34]: just, to ask, how do you define world model?Swyx [00:48:38]: Oh, yeah, let's go there.Ethan [00:48:40]: SoVibhu [00:48:40]: So just for context, we talked about, video generation, and then there's a-- if you say there's a distinction between world models, what's your, what's your definition? How do you see the two?Ethan [00:48:53]: So disclaimer, I'm not going to debate, what is world model. Yeah. there are many definitions, so I'll just talk about my definition. Since I came from the multi-model, multi-model domain, so mainly talking from video. So world model is like real-time interactive long horizon videos. So there are three parts. so we-- let's talk about them one by one. So the so interaction, so we just, we just look at Facebook and neural computer. So the interaction part of it, so you, world model can allow you to interact with them through keyboard, mouse, and maybe also voice. So these all is-- all is a modality. You can, you can interact with the model, and the model should respond reasonably. Second part is real time. So once you, once, say, you move your mouse, if, say, the world model generate a game, how fast can the game respond? So if you're like professional CS: GO players- -my say, oh, you have to respond- He's beginner within sub ten milliseconds or- Yeah even less. So that's not most of the- No, sixty FPS. Let's go. Oh, three hundred FPS. Oh, five hundred FPS. Wait. okay, yeah. I didn't do the math, but yeah, okay. Uh- Yeah, three hundred FPS, that's a three millisecond. So you have to respond- Oh, s**t. Okay. YeahEthan [00:50:29]: within a millisecond. Most of the video models cannot do that. Yeah. And, but if you, say, if you have a video model that is, say, like a digital human, the response time might be more generous. Maybe typically, for real-time voice interaction, it's like two hundred millisecond. So that's, that's much more generous. But even two hundred millisecond is pretty, it is pretty tricky, ‘cause remember we mentionedEthan [00:51:01]: you have this, temporal compression coming from the VAE. So if you, if you don't compress the temporal dimension, your sequence length is going to explode. So if you want to have this real-time, real-timeness in your model, you have to do is one context problem. And the third part is long horizon, ‘cause we-- if you're not going to just play with, video games just, a few seconds, most video models only a few seconds. We're going to play with minutes, hours. The model have to be able to generate long-form content.Ethan [00:51:42]: So putting these three together, it's, real-time, long horizon interactive videos. I think the final state will be, for example, like a video, a video version of Playbook, where you can, you can interact with, a neural computer. You move your mouse, and you click on the generative interface, and it will reply to you through pixels- generating in real time. But getting there, it's, it's a very long way to get there. So one of the first step, at Grok Imagine, where I led a small world model team there, was to build video extension. So, video extension- it's the first step of interactivity. Yeah. It's, it's the first step. Yeah. So it's the first step- You have it here, video editing, yeah. Yeah. Yeah. So the first step is because, this unlocks long horizon videos. Typically, for most of the video generation models, you give it a prompt or an image as an initial frame. You generate video, that's it. That's just, one time, done. And some creators would try to, use the last frame as a first frame for the second video. It can-- sometimes it works, but if you do it a few times, it says the quality would decrease. And- It doesn't have that context- Yeah over the full video, so the temporal- Yeah, exactly. Yeah, ‘cause you only gave it the last frame, of course, right? Yeah. Exactly. And- it's actually a pretty fun hack. if you've seen like- Oh, no, he's saying something better. Yeah. And for example, like Vue, I remember Vue 3 has like a second context of the last video. It is slightly better than using the last frame, but it has the same problem-- similar problem that it, the quality would decrease. if you extend a few times to, one minute, the video quality would look much worse than the first video. Second, another problem is that the model doesn't have long-range knowledge of, what's happening before. Say, if they generate some dialogue, some, two people speaking, and their voice might change, over some time, especially if the second conditioning, it does not cover the previous context. So these are the core challenges. So the Grok Imagine video extension, it has historical context of all of the previous generated videos. It can, It has, it has the context of, who is speaking and what objects have appeared and everything, having that to generate the next video. So if we naively do this, you can imagine, just, put all of the previous history video tokens into the context. The context lens will easily explode. Especially for video models, that can be like a few, a few million context, I would imagine- context lens. Yes.Yeah.Swyx [00:54:58]: Let's run with that.Ethan [00:54:59]: for example, like in Cosmos, I think just five seconds of video is like a fifty K or sixty K number of tokens. So like if you do, if you do fifty second, that's a five hundred K tokens. If you do longer than that, easily explode. This long horizon, problem was the first step we're trying to solve world model. It turns out people, yeah, people love video extension. Like a lot, a lot of the creators love using video extension to create longer form videos. This is the part I liked that you have a, you have an intermediate step toward the final goal instead of just a straight shot to the final version very much.Swyx [00:55:48]: But I can see you have a strong vision of where we want to end up.Long Context, Redundancy, and Efficient Interactive VideoVibhu [00:55:51]: Does it seem like it's an efficiency issue? okay, we're at a few million tokens context,. If you draw the parallel to language models, we had very short context, two thousand, eight thousand, then, you scale it up one million, ten million. sure, there's effective context, but at the end of the day, it's just what's it worth? sure, there's a whole training data side. In video, it might be slightly easier ‘cause we have a hundred million token video, right? Just take a movie with the full context there. Like is this efficiency from an inference standpoint that like it's expensive, but we know how to solve it? Or like why is this not the approach? So like my broader point was on your second point of world models, you say it needs to be interactive and live, right? You should be able to play a game and see the interaction live. So one thing I see with research is a lot of what you actually serve is different than what you build, right? So we talked about distillation. You train big model, you distill it, you do quantization, speculative decoding. We do all this stuff to serve it efficiently. Should we not just have a solution, like a world model that can interact well, do inference optimization, serve it, distill it secondary, so make it real time after you solve it? So like a-- another parallel is say, continual learning, right? What we need is someone to solve it and show it works inefficiently. Give it a few years, people will make it efficient. Same thing with regular attention, right? It worked. Over a few years, people have different forms of attention, and we've scaled it to be efficient at log context,? So kind of two things there, right? One is it seems like it works. You've scaled it. Can we not just scale it a lot more efficiently over time? Do we need a separate approach if this works? And same thing with interaction, right? if we can get it done, like if we can solve some way that it works, we can solve making it more efficient from an inference standpoint later.Ethan [00:57:53]: that's actually a very good point. So in videos, there's actually a lot of redundancies. So we solve a lot of the pixel redundancy from VE, but there's more redundancy in long range and long horizon videos. Say, if a character appear in the first clip and then it disappeared, it only reappear at the end of the video, you probably don't need the-- the context, like in the middle of the generation. So you only need that character, where you need. So that's why, I helped build another feature. It's a reference video.Vibhu [00:58:36]: Is it here?Swyx [00:58:36]: is it the same model release or different one?Ethan [00:58:39]: It's a different one.Ethan [00:58:41]: You probably need to search onSwyx [00:58:43]: I'll find itEthan [00:58:43]: X reference to video.Ethan [00:58:46]: So reference video allow you to like upload up to seven images as condition and generate the video. Say, if like I want-- it can, it can be characters or objects or even scenes. Say like I want, I want condition on, Sean's selfie and holding a bladeSwyx [00:59:07]: We have a dogEthan [00:59:08]: or whatever.Swyx [00:59:08]: We put the dog in the thing.Ethan [00:59:09]: you can put them there and the video models will generate the video from and copies the context over. So that can solve a lot of the problems there, like the long context problem. It doesn't need to have a very long context, but it's-- I feel like it's an intermediate solution. The modelSwyx [00:59:29]: It's cheating.Ethan [00:59:30]: the model should be able to like selectively know, where should I draw the references. So say if I want to generate a movie, I generate it autoregressive, like a ten second at a time or something. And now this character appear, I can look back to where it first appear and, bring that back. Yeah, this one, I put the references. Yeah, that's, Optimus, Einstein myself, Annie.Vibhu [01:00:02]: Oddly enough, I used Grok Search to find it, and it pulled your LinkedIn post. But yeah we found it.Ethan [01:00:08]: Interesting.Vibhu [01:00:10]: ButxAI's Underrated Work, Culture, and WatermarkingSwyx [01:00:11]: this is a problem. This is not your fault, but like XAI doesn't communicate all this work that you do very well because they just have the model release and then that's it. But actually, these details are very good.Swyx [01:00:22]: As far as I understand, everything you just described is state-art, like no one else has done it.Vibhu [01:00:30]: A lot of-- yeah, I have a lot moreSwyx [01:00:32]: And then, and then you just put this blog post with the cookies. I'm this is not enough,?Swyx [01:00:37]: but I, obviously this is like the high level numbers that people want to know. But no, okay, soVibhu [01:00:42]: And I wonder, like part of that is also some labs don't share research into what happens. And ifSwyx [01:00:50]: No, but this is literally bragging about how good they are, right?Swyx [01:00:54]: Like, why would you not say that you are capable of extending with full context? this is not a secret sauce. This is like we did the work. yeah, I don't know.Ethan [01:01:02]: different labs have slightly different communication styles.Swyx [01:01:07]: Anyway, if anyone from XAI is listening we are always happy to help you tell your story. Yeah, okay, so you did references, and I think, I think kind of the point you're, you're making is it is sort of like a kludge, right? this is-- you can do seven, but what about 100?Swyx [01:01:23]: Right? Then you need a completely different thing.Ethan [01:01:26]: So I think it's-- this is, a mechanism to, select the context from the history, and you might not put the entire history into the context. for example, there's a paper called Frame Pack, which haveEthan [01:01:41]: a heuristic that the latest history, the last one second, I put the entire history, and the history before that, I would, compress it and makes the video smaller. So they follow this pattern, this build overall pattern that the maximum sequence length is fixed. So the further you are from the current frame, you have a smaller image. So this is just a heuristic. I think it can be more automatic. The model is aware like which history part of it can be select. So this part of the research is actually being actively, worked on by a lot of people. It's also quite interesting. I feel this is actually, this part of long context is a little bit ahead of the LLM part.Ethan [01:02:31]: So for example, like in LLMs, if you-- so contexts keep growing. Let's say if you call tool and the tool call history is extremely long, that's still in context, and keep growing, keep growing. Even if you switch the topic to something else, the whole context was there. There are some agentic harnesses that help you to, say, prune the tool results and, prune Like when you, when you query a file, only show like the top 200 lines or something. Those were very heuristic-driven.Swyx [01:03:08]: For listeners, we did a write-up on the cloud code, leak where there are eight different kinds of pruning, including like you prune the tool results and all that. So you can, you can read up on that kind of thing.Ethan [01:03:17]: I think, one breakthrough in continual learning might be like a way to automatically, manage its own context.Swyx [01:03:27]: These are all heuristics, and they will be replaced by machine learning.Ethan [01:03:30]: InterestinglyVibhu [01:03:32]: TheEthan [01:03:32]: the same thing is being researched in both LLMs and video models.Vibhu [01:03:36]: The interesting thing is also like in the paper you showed, it's actually happening at the model level, right? Compared to like language models, sure, we have base attention, but we'll do our own compression, we'll do our own pruning, which is separate from model error.Vibhu [01:03:49]: Eventually, it all just boils in, hopefully.Swyx [01:03:52]: I think this is a form of like attention, but like also know sort of reasoning attention. I feel like that's different than normal attention.Swyx [01:04:03]: Does that, does that make sense?Ethan [01:04:04]: It's, it's different in the sense that attention, not to mention, set sparse attention aside,

This summer, HRP is reading Pedagogies of Collapse: A Hopeful Education for the End of the World As We Know It, by Ginie Servant-Miklos, and we're inviting you to join us. Visit humanrestorationproject.org/book-club to sign up for our summer book club, where we'll meet to discuss the ideas and implications of Pedagogies of Collapse and be joined by the author, for a Q&A on July 31. I'll include a link to the book in the show notes, which is available on Open Access through Bloomsbury. Hope to see you there!I'm back this week with another narrated piece from our upcoming Progressive Education Primer. If you like this format and want to have more narrated essay content, or if you can't stand it, leave a comment on YouTube or Discord to let us know. This one is written by our Executive Director, Chris McNutt, titled Teaching in the Wreckage of the Real.HRP Book ClubPedagogies of Collapse, Bloomsbury Open AccessTeaching in the Wreckage of the Real, Chris McNuttAdditional music credits: Dandelion by | e s c p | https://www.escp.space | https://escp-music.bandcamp.com

What's the hidden tax your organization pays every time a creative asset moves from a design tool to a marketing platform, and how can you shorten the time to gain important insights about how your campaigns perform?Agility requires more than just speed. It demands that we eliminate the friction between our systems and processes so teams can move from concept to customer with minimal translation errors and maximum impact. It also means that we need to find the best ways to understand campaign performance without requiring everyone in marketing to be a data scientist.We're going to discuss:- the persistent gap between creative design and marketing execution- the value that AI-based capabilities can add to the understanding of analytics and performanceTo help me discuss this topic, I'd like to welcome Ose Amiegheme, Head of Email Product at Intuit Mailchimp. About Ose Amiegheme Ose Amiegheme is a product leader building the future of creation and growth tools.Today, he leads product for Intuit Mailchimp's Email and omnichannel campaigns creation experiences, shaping how small businesses create content, launch campaigns, and grow across channels.Previously, he led advertising products at TikTok supporting multi-billion-dollar revenue businesses and helped launch products spanning GenAI creative tooling, campaign optimization, and advertiser control systems.Before TikTok, Ose spent four years at Adobe helping build Adobe Express, where he worked across editor experiences, AI-assisted creation, and products used by millions of creators globally. His career has followed a consistent theme of building products that empower creators and marketers to tell their story in a way that feels genuine but also standout.Outside of work, Ose is a huge soccer fan and he is excited for the upcoming soccer World Cup. Ose Amiegheme on LinkedIn: https://www.linkedin.com/in/ose-amiegheme/ / https://www.linkedin.com/in/jeremyejones/ ---------- Resources ---------- Intuit Mailchimp: https://mailchimp.com/ The Agile Brand podcast is brought to you by TEKsystems. Learn more here: https://aglbrnd.co/r/2868abd8085a9703 Drive your customers to new horizons at the premier retail event of the year for Retail and Brand marketers. Learn more at CRMC 2026, June 1-3. https://aglbrnd.co/r/d15ec37a537c0d74 We're proud to be a media partner for #MAICON26 - Oct. 13-15! Learn how AI can power your marketing and business and help you grow smarter. Use code AGILE150 to save! https://aglbrnd.co/r/7fe458ced0f04658Reach your customers with Reddit. Spend $500 in ad spend, get $500 back in ad credit! Learn more: https://advertalize.com/r/491818c79fb1873fDon't miss We Make Future - the International Festival of Innovation in AI, Tech, and Digital Marketing, June 24-26 in Bologna. Learn more: https://aglbrnd.co/r/c80991afff416bb2The most influential minds in software, AI, and engineering leadership will be at WeAreDevelopers World Congress North America, September 23-25 in San Jose. Learn more: https://aglbrnd.co/r/60a7299222a7bcf1 Enjoyed the show? Tell us more at and give us a rating so others can find the show at: https://aglbrnd.co/r/faaed112fc9887f3 Connect with Greg on LinkedIn: https://www.linkedin.com/in/gregkihlstromDon't miss a thing: get the latest episodes, sign up for our newsletter and more: https://aglbrnd.co/r/35ded3ccfb6716ba Check out The Agile Brand Guide website with articles, insights, and Martechipedia, the wiki for marketing technology: https://www.agilebrandguide.com The Agile Brand is produced by Missing Link—a Latina-owned strategy-driven, creatively fueled production co-op. From ideation to creation, they craft human connections through intelligent, engaging and informative content. https://www.missinglink.company Hosted on Acast. See acast.com/privacy for more information.

You thought the Apple Vision Pro was expensive, but now you could have to choose between buying 180 of the headset, or one Ferrari Luce designed by Jony Ive. Or you could just enjoy the good, the bad, and the sometimes silly iPhone rumors that came out this week, on the AppleInsider Podcast.Contact your hosts:@williamgallagher_ on Threads@WGallagher on TwitterWilliam's 58keys on YouTubeWilliam Gallagher on emailWes on BlueskyWes Hilliard on emailWes's blog HillitechSponsored by:MasterClass: Get 15% off annual memberships at MasterClass.comNordStellar: Unlock your 10% discount at nordstellar.com/appleinsider with the coupon code nordappleinsider-10-NORDSTELLARLinks from the Show:iPhone 18 color 'leak' from fake account appears to be camera protector, not componentiPhone 18 clear cases could revert to old MagSafe design for some reasonManufacturers are taking a big chance on iPhone Fold case listingsRumored anti-snatch feature will automatically lock iPhones yanked out of a user's handThis is what the Siri redesign might look like in iOS 27'GenAI' Apple subdomain surfaces weeks ahead of WWDCApple's worst AI feature to get a 'big boost' with upgraded Apple Foundation ModelsFormer Apple designer's take on Ferrari will upset fans of the vehicle brandFuture iPhone might get real underwater photography featuresApple Vision Pro & PlayStation 5 are the perfect combo with Portal Remote Play appSupport the show:Support the show on Patreon or Apple Podcasts to get ad-free episodes every week, access to our private Discord channel, and early release of the show! We would also appreciate a 5-star rating and review in Apple PodcastsMore AppleInsider podcastsTune in to our HomeKit Insider podcast covering the latest news, products, apps and everything HomeKit related. Subscribe in Apple Podcasts, Overcast, or just search for HomeKit Insider wherever you get your podcasts.Subscribe and listen to our AppleInsider Daily podcast for the latest Apple news Monday through Friday. You can find it on Apple Podcasts, Overcast, or anywhere you listen to podcasts.Those interested in sponsoring the show can reach out to us at: advertising@appleinsider.com (00:00) - Intro (01:45) - Silly leaks (22:10) - Good leaks (31:56) - GenAI (53:20) - Ferrari Luce (01:00:19) - Shot on iPhone (01:10:56) - Apple Vision Pro gamining ★ Support this podcast on Patreon ★

AI feedback drives continuous improvement in Gen AI tools by helping organizations adapt faster, improve user satisfaction, boost efficiency, and create a culture where employee insights lead to smarter innovation and better results. That's the key take-away message of this episode of the Wise Decision Maker Show, which talks about why Gen AI feedback is the real advantage.This article forms the basis for this episode: https://disasteravoidanceexperts.com/forget-algorithms-gen-ai-feedback-is-the-real-advantage/

On this episode, we explore what rigorous AI safety testing looks like for customer-facing AI — and why most deployments carry more risk than the teams running them expect.Testing AI before launch is standard practice. But one-time manual testing treats AI like a deterministic system. Model behavior is probabilistic, and the consequences of inadequate testing fall into four categories: people data harm, other types of data harm, reputational harm and commercial harm. Each represents a distinct exposure with real consequences for your organization and your customers.Meaningful AI safety testing requires something different: continuous, automated adversarial testing at scale, designed to find what a bad actor would find before they find it. TELUS Digital's benchmark research, running 34 models through more than 620,000 simulated attacks, found attack success rates ranging from 1% to 90%, and identified five gaps in how most organizations approach testing: scale, scope, variety, repetition and simulation realism.Bret Kinsella, senior vice president and general manager of Fuel iX at TELUS Digital, draws on the GenAI safety model benchmark report to explain where CX AI tends to fail adversarial testing methods, how the exposure management framework reframes risk as an ongoing operational discipline and the question every CX leader should be asking about the AI their customers are currently interacting with.Show notesWatch Uncharted, TELUS Digital's AI safety and security summit, on demand: https://www.telusdigital.com/insights/fuel-ix/resource/uncharted-ai-security-safety-summit-videosDownload the full GenAI safety model benchmark report: https://www.telusdigital.com/insights/fuel-ix/resource/genai-safety-benchmark-2026Learn more about Fuel iX Fortify, TELUS Digital's continuous adversarial testing and validation platform for enterprise AI, and request a free AI safety & security analysis: https://www.telusdigital.com/solutions/fuel-ix/fortifyConnect with Bret Kinsella on LinkedIn: https://www.linkedin.com/in/bretkinsella/

Show Notes & References Resources mentioned in this episode: Tither, E. (2025, December 10). ⁠⁠What happens to the data you feed LLMs?⁠⁠ University of Illinois System, Student Money Management Center. https://blogs.uofi.uillinois.edu/view/7550/1055573584 Chen, K., Zhou, X., Lin, Y., Feng, S., Shen, L., & Wu, P. (2025). A survey on privacy risks and protection in large language models. ⁠⁠Journal of King Saud University – Computer and Information Sciences⁠⁠, ⁠⁠37⁠⁠(7). https://doi.org/10.1007/s44443-025-00177-1 Farooqui, A. (2025, February 12). ⁠⁠Samsung lets employees use ChatGPT again after secret data leak in 2023⁠⁠. SamMobile. https://www.sammobile.com/news/samsung-lets-employees-use-chatgpt-again-after-secret-data-leak-in-2023/ Han, X., Peng, H., & Liu, M. (2025). The impact of GenAI on learning outcomes: A systematic review and meta-analysis of experimental studies. ⁠⁠Educational Research Review⁠⁠, 100714. https://doi.org/10.1016/j.edurev.2025.100714 Imperial War Museums. (2018). ⁠⁠How Alan Turing cracked the enigma code⁠⁠. https://www.iwm.org.uk/history/how-alan-turing-cracked-the-enigma-code Kwak, R. (2023, November 30). ⁠⁠Announcing Microsoft Copilot with Data Protection⁠⁠. Technology Services, University of Illinois. https://www.techservices.illinois.edu/2023/11/30/announcing-microsoft-copilot-with-data-protection/ Kwak, R. (2025, November 11). ⁠⁠ChatGPT arrives at U of I⁠⁠. Technology Services, University of Illinois. https://www.techservices.illinois.edu/2025/11/11/chatgpt-arrives-at-u-of-i/ Microsoft 365, Copilot with Data Protection – AI Chat for the Web. (2024). University of Illinois System KnowledgeBase. https://answers.uillinois.edu/133037 OpenAI. (2023). ⁠⁠Privacy policy⁠⁠. https://openai.com/en-GB/policies/row-privacy-policy/ Ray, S. (2023, May 2). ⁠⁠Samsung bans ChatGPT among employees after sensitive code leak⁠⁠. Forbes. https://www.forbes.com/sites/siladityaray/2023/05/02/samsung-bans-chatgpt-and-other-chatbots-for-employees-after-sensitive-code-leak/ Yao, Y., Duan, J., Xu, K., Cai, Y., Sun, Z., & Zhang, Y. (2024). A survey on large language model (LLM) security and privacy: The good, the bad, and the ugly. ⁠⁠High-Confidence Computing⁠⁠, ⁠⁠4⁠⁠(2), 100211. https://doi.org/10.1016/j.hcc.2024.100211

In this episode of The Good Life EDU Podcast, Andrew Easton welcomes back Amanda Bickerstaff, CEO and Co-Founder of AI for Education, for a timely conversation about where schools are in their generative AI journey and what practical next steps can help educators, leaders, and students move forward with greater clarity. Amanda shares the thinking behind AI for Education's new SEE Framework, which defines generative AI literacy as the knowledge, mindsets, and practices that allow individuals to use GenAI safely, ethically, and effectively. The conversation explores why AI literacy cannot be limited to teachers or leaders alone, why schools should avoid gatekeeping these conversations from students, and how clear guardrails can help learning communities make more intentional decisions about AI use. Andrew and Amanda also discuss the practical dilemmas schools are facing right now, including AI-generated feedback, AI grading, AI detection, student use of chatbots, Google AI Overviews, and the need to move beyond simple “AI is good” or “AI is bad” binaries. Amanda emphasizes that the work is not about adopting every AI tool or avoiding AI altogether, but about building the judgment to know when AI use is appropriate, when it is not, and how to remain intentional, critical, transparent, responsible, and committed to continued learning. This episode also previews Amanda's keynote and breakout session at the Future Ready Nebraska Conference, where she will explore the SEE Framework and provide practical examples, including “SEE snapshots,” that help educators think through real-world AI decisions using the framework's lenses of safe, ethical, and effective use. Resources: AI for Education Resource hub for AI literacy, guidance, prompt libraries, webinars, and practical supports for schools. https://www.aiforeducation.io/ The SEE Framework: A Practical Guide to Building Generative AI Literacy AI for Education's generative AI literacy framework focused on the knowledge, mindsets, and practices needed to use GenAI safely, ethically, and effectively. https://www.aiforeducation.io/the-see-framework-for-generative-ai-literacy AI Course for Educators — AI for Education A free, self-paced, two-hour course designed to help educators get started with AI, developing prompting strategies, considering ethical implications, and introducing AI to students responsibly. https://www.aiforeducation.io/ai-course Beyond the AI Inflection Point A futurist piece from AI for Education that explores a fictional school district's AI journey from 2023 to 2030 and invites educators to think through the choices schools are facing now. https://www.beyondtheaiinflectionpoint.com/ Future Ready Nebraska Conference Amanda Bickerstaff's keynote and breakout session will highlight the SEE Framework and practical AI literacy decisions for schools. https://nefutureready.com/

Realities Remixed, formerly known as Cloud Realities, launches a new season exploring the intersection of people, culture, industry and tech.Today's most pressing challenges arise from the collision of rapid technological change with deepening economic inequality, weakening democratic systems, geopolitical instability and accelerating climate pressure, leaving world leaders wrestling with how to govern and solve these deeply interconnected crises.This week, Dave, Esmee and Rob are joined by Dex Hunter-Torricke, Founder & President The Center for Tomorrow to explore how tech can solve world macro issues. TLDR00:33 – Introduction00:40 – Hang out: The Boys on Amazon Prime final episode (spoilers) 06:02 – Dig in: How to solve world macro issues? 07:45 – Conversation with Dex Hunter-Torricke 44:52 – Writing a book and meeting world leaders GuestDex Hunter-Torricke: https://www.linkedin.com/in/dextb/https://www.centerfortomorrow.com/ HostsDave Chapman:  https://www.linkedin.com/in/chapmandr/Esmee van de Giessen:  https://www.linkedin.com/in/esmeevandegiessen/Rob Kernahan:  https://www.linkedin.com/in/rob-kernahan/ ProductionMarcel van der Burg:  https://www.linkedin.com/in/marcel-vd-burg/Dave Chapman:  https://www.linkedin.com/in/chapmandr/ SoundBen Corbett:  https://www.linkedin.com/in/ben-corbett-3b6a11135/Louis Corbett:   https://www.linkedin.com/in/louis-corbett-087250264/ 'Realities Remixed' is an original podcast from Capgemini

Haroon Inam is Co-founder and CEO of DG Matrix, a company that makes the world's most compact Power Router, aggregating distributed energy for GenAI datacenters, microgrids, fleet electrification, and associated systems. As AI workloads drive unprecedented electricity demand and legacy grid infrastructure struggles to keep pace, DG Matrix has commercialized the world's first multi-port solid-state transformer to meet the energy needs. In this episode, Inam explains why transformer bottlenecks, distributed generation, and 800V DC architectures are reshaping the future of power delivery for AI infrastructure. He discusses DG Matrix's product strategy, manufacturing scale-up plans, and the role of software-defined power systems in next-generation data centers. Finally, Inam shares his take on the future of distributed microgrids and “cellular power” and how to scale power electronics manufacturing. DG Matrix recently closed a $60 million Series A led by Engine Ventures that MCJ is proud to have participated in.  Episode recorded on May 13, 2026 (Published on May 26, 2026) In this episode, we cover:  (00:00) Overview of DG Matrix  (01:41) Introducing the Founders: Haroon Inam and Dr. Bhattacharya (05:25) How traditional grid architecture became constrained for AI workloads (09:57) Solid-state transformers (SST), multi-port systems and voltage classes (12:18) Why early SST efforts struggled economically (13:13) How DG Matrix's multi-port architecture works (16:48) Comparing DG Matrix hardware footprint to legacy power systems (20:08) Transformer shortages and data center infrastructure bottlenecks (24:27) DG Matrix's medium-voltage and low-voltage product strategies (27:55) Product rebranding and current commercial deployments (30:45) Partnerships with EPC firms, battery providers, and turbine manufacturers (34:27) Manufacturing scale-up plan and hyperscaling production (36:36) Supply chain strategy to avoid rare earth dependencies (38:16) Reliability engineering and software-defined power systems (43:47) DG Matrix's go-to-market and hybrid hardware/software business model (46:36) The vision for distributed “cellular power” (48:14) Utilities, microgrids, and the future of interconnected distributed infrastructure Enjoyed this episode? Please leave us a review! Share feedback or suggest future topics and guests at info@mcj.vc.Connect with MCJ:Cody Simms on LinkedInVisit mcj.vcSubscribe to the MCJ Newsletter*Editing and post-production work for this episode was provided by The Podcast Consultant

We showcase recordings from this year's RSAC. At RSAC Conference 2026, Scott Clinton, Co-Chair and co-founder of the OWASP GenAI Security Project, shares insights from the project's latest research, including new landscape guides and evolving approaches to securing generative and agentic AI systems. The conversation explores critical gaps in GenAI data security, the rise of AI-assisted development, and the immense growth of the OWASP community and sponsor ecosystem. Looking ahead, he outlines the most urgent risks and priorities shaping AI and agentic security in 2026. Then Merritt Maxim discusses how AI is affecting Identity and Access Management. Expect to hear this topic a lot throughout 2026, especially as the industry tries to figure out what's different or special about securing agent identities. We close with a chat with Janet Worthington about the impact of agents on the SDLC and how orgs are updating their controls to deal with code generated by humans and LLMs alike. Segment Resources: https://genai.owasp.org https://genai.owasp.org/resources/ https://www.scworld.com/podcast-episode/3905-keeping-up-with-the-owasp-genai-project-scott-clinton-asw-381 This segment is sponsored by The OWASP GenAI Security Project. Visit https://securityweekly.com/owasp to learn more about them! Visit https://www.securityweekly.com/asw for all the latest episodes! Show Notes: https://securityweekly.com/asw-384

What happens when you stop focusing on human resources and start focusing on the human experience? Technology is advancing faster than human nature can keep up. If you want to stay relevant, you need a fundamental cultural reset, not just new software.  In this interview, Laura Cushing, the Chief People Experience Officer at Pacific Life, discusses the evolving intersection of organizational culture, employee engagement, and artificial intelligence. She emphasizes a strategic shift toward accountability and transparency as the balance of power moves back toward employers in a post-pandemic landscape. To prepare for an AI-driven future, Pacific Life has implemented a Gen AI Academy and Innovation Labs to demystify technology and help staff reimagine their workflows. Cushing highlights the rising importance of "power skills"—human-centric abilities like coaching and visionary leadership—which remain essential as technical tasks become automated. Ultimately, she argues that HR leaders must cultivate deep business acumen and proactive trust-building to successfully guide their workforces through digital transformation. Watch the full video on YouTube ---------- Start your day with the world's top leaders by joining thousands of others at Great Leadership on Substack. Just enter your email: ⁠⁠https://greatleadership.substack.com/ Quick heads-up: my new book, The 8 Laws of Employee Experience, is a practical playbook for building an environment where people do their best work—order a copy here: https://bit.ly/8exlaws

How Deeply Human Is Language? Chomsky, the Brain, and the AI Fantasy (MIT Press, 2026) is Yosef Grodzinsky's exploration of the criticality of the linguistic theories to the design of LLMs. The book dwells on the significance of the marriage between computational and theoretical fields, specifically “engineering and science” on the development of unique Language Learning Models. Yosef maintains that leveraging linguistic theories for the development of Gen AI chatbots and training of Language Learning Models will help the growing Gen-AI revolution. In the book, LLMs are evaluated from the neurolinguistic perspective, comparing how the human brain works with different LLMs' reactions to prompts, highlighting how a collaboration between the core linguists and the experts in the technology-related fields could make a change. Yosef Grodzinzky's positions in the book is grounded in contemporary linguistics, founded and inspired by Noam Chomsky, the father of the “mentalist” linguistic perspective to language acquisition. In the book, the author employs the historical approach to tell different significant stories to communicate multiple messages of success of interdisciplinary practices. While the main idea is to explore the centrality of linguistic science to other fields with specific emphasis on Engineering and sister's technological fields, the book dwelled on specific pitfalls of the linguistics and way forward to promote novel interdisciplinary productions. Mariam Olugbodi is a university teacher and a writer, she is the author of the monograph titled: “Stylistic Features in the 2011 and 2012 Final Matches Commentaries in the UEFA Champions League”, published by Grin Verlag. Mariam's greatest dream is seeing a world where knowledge is accessible to all. She does this through her volunteering roles on open knowledge platforms as a host and an editor. As part of her effort to maintain inclusion and diversity in knowledge transmission, she volunteers as a teacher in crises contexts. Learn more and connect with Mariam through her social links here. | LinkedIn| here. |ORCID| and here. |Meta| 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