Podcasts about Omni

Darrach Ó Duibh has such an interesting story, and I wanted to speak with him to discuss his incredible life journey and the profound, unexpected insights he gained while translating the Book of Mormon into the Irish language longhand. He's an ancient language researcher, polyglot, and textual critic based in Ireland. Having previously worked as a bibliographic researcher for FARMS (now the Maxwell Institute) under the legendary LDS scholar Hugh Nibley, Darrach started this translation project 20 years ago while bedridden after a honeymoon car crash. Adding another dimension to his life, Darrach is also a well-respected professional mixed martial arts (MMA) referee.We explore his unique personal interactions with Hugh Nibley, his background growing up with a father in military intelligence, and the powerful textual evidence for multiple, distinct prophetic voices that only becomes visible when translating the Book of Mormon text word by word.Some highlights from this episode include:Working with Hugh Nibley: Darrach shares first hand stories from his time sitting across the desk from Hugh Nibley at BYU, including Nibley's shocking command of ancient languages, his quiet patience with students, and a legendary family story involving a moving train in Utah.Growing Up in Military Intelligence: From learning languages at six years old to witnessing his father barter with guards in Russian at Checkpoint Charlie, Darrach explains how his childhood in Germany unlocked his lifelong love for language.The Ancient Egyptian Pattern in the Text: A fascinating deep dive into the very first verses of First Nephi, Enos, Omni, Mormon, and Moroni, revealing a complex, authentic ancient narrative pattern known as the narrative infinitive or "stacking" that Joseph Smith could not have easily fabricated.Hearing Distinct Voices: Darrach details how his translation work forced him to recognize completely different linguistic registers, vocabulary, and tones between Nephi, Jacob, and Mormon, providing stunning textual evidence for the book's authenticity.Faith, Revelation, and Suspension: A discussion on why the Lord purposely keeps archaeological and metaphysical proof in a state of suspension, allowing spiritual knowledge and true faith to flourish rather than letting it become dormant.You can find Darrach's Irish translation of the Book of Mormon at the following link:Amazon: An Leabhar Mhórmoin: Fianaise eile ar Íosa Críost (Irish Edition) https://www.amazon.com/Leabhar-Mh%C3%B3rmoin-Fianaise-Cr%C3%ADost-Irish/dp/B0H2W35RW4/Follow For All The Saints on social media for updates and inspiring content:www.instagram.com/forallthesaintspodhttps://www.facebook.com/forallthesaintspod/For All The Saints episodes are released every Monday on YouTube, Spotify, Apple Podcasts and more:https://www.youtube.com/watch?v=TVDUQg_qZIU&list=UULFFf7vzrJ2LNWmp1Kl-c6K9Qhttps://open.spotify.com/show/3j64txm9qbGVVZOM48P4HS?si=bb31d048e05141f2https://podcasts.apple.com/gb/podcast/for-all-the-saints/id1703815271If you have feedback or any suggestions for topics or guests, connect with Ben & Sean via hello@forallthesaints.org or DM on InstagramConversations to Refresh Your Faith.For All The Saints podcast was established in 2023 by Ben Hancock to express his passion and desire for more dialogue around faith, religious belief, and believers' perspectives on the topics of our day. Tune into For All The Saints every Monday on YouTube, Spotify, Apple Podcasts, and more.Follow For All The Saints on social media for daily inspiration.

This week on the Drive Thru, Jim looks at the top wrestlers in their 40s in 1984! Plus Jim answers YOUR questions about Sol Ruca, Ethan Page, Paul E. Dangerously's phone, The Omni, Triple H as a booker, Sting, Andre The Giant tagging with Bobby Eaton, the announcement of Kurt Angle's TNA signing, and much more! Also, Jim looks at an issue of Matwatch from March 1990! Thanks to our episode sponsors: RAYCON: Upgrade your dad’s everyday routine. Go to buyraycon.com/jce to get 15% off. Thanks Raycon for sponsoring! HEXCLAD: Find your forever cookware @hexclad and get 10% off at hexclad.com/JCE! #hexcladpartner Send in your question for the Drive-Thru to: CornyDriveThru@gmail.com Follow Jim and Brian on Twitter: @TheJimCornette @GreatBrianLast Merch! https://arcadianvanguard.com/ Join Jim Cornette's College Of Wrestling Knowledge on Patreon to access the archives & more! https://www.patreon.com/Cornette Subscribe to the Official Jim Cornette channel on YouTube! http://www.youtube.com/c/OfficialJimCornette Visit Jim's official site at www.JimCornette.com for merch, live dates, commentaries and more! You can listen to Brian on the 6:05 Superpodcast at 605pod.com or wherever you find your favorite podcasts!See omnystudio.com/listener for privacy information.

This week on the Drive Thru, Jim looks at the top wrestlers in their 40s in 1984! Plus Jim answers YOUR questions about Sol Ruca, Ethan Page, Paul E. Dangerously's phone, The Omni, Triple H as a booker, Sting, Andre The Giant tagging with Bobby Eaton, the announcement of Kurt Angle's TNA signing, and much more! Also, Jim looks at an issue of Matwatch from March 1990! Thanks to our episode sponsors: RAYCON: Upgrade your dad’s everyday routine. Go to buyraycon.com/jce to get 15% off. Thanks Raycon for sponsoring! HEXCLAD: Find your forever cookware @hexclad and get 10% off at hexclad.com/JCE! #hexcladpartner Send in your question for the Drive-Thru to: CornyDriveThru@gmail.com Follow Jim and Brian on Twitter: @TheJimCornette @GreatBrianLast Merch! https://arcadianvanguard.com/ Join Jim Cornette's College Of Wrestling Knowledge on Patreon to access the archives & more! https://www.patreon.com/Cornette Subscribe to the Official Jim Cornette channel on YouTube! http://www.youtube.com/c/OfficialJimCornette Visit Jim's official site at www.JimCornette.com for merch, live dates, commentaries and more! You can listen to Brian on the 6:05 Superpodcast at 605pod.com or wherever you find your favorite podcasts!See omnystudio.com/listener for privacy information.

In Part 2 of this two-part podcast series, Rebecca and Vickie continue discussing their Florida educational tour on behalf of their travel agency. Building on the experiences shared in Part 1, they dive deeper into the highlights of the Disney Cruise ship and the Omni Convention Center, offering additional insights into guest experiences, event planning opportunities, and travel recommendations. They reflect on key takeaways from the tour, discuss how these experiences will benefit their clients, and share valuable tips for travelers considering a Disney cruise, group event, or Florida getaway. Follow us on all our social media accounts on Facebook and on Twitter at @Mousecapadespod. Thinking about being a guest on our show, or have a question or comment? Contact us anytime via text or phone at 636-373-4497. Have a magical day my friends!

Full show notes and transcript  - https://bit.ly/google-agentic-eraWatch on YouTube - https://youtu.be/eamMBmm6oTU-----Episode Summary:Dara and Matthew open with a breaking-news bulletin on Anthropic's newly released Fable, the consumer sibling to Mythos, covering its safety off-ramp to Opus 4.8, its pricing, and the looming switch from subscription to usage-based access. The main episode is a deep dive on Google Cloud Next '26 and I/O '26, unpacking the Gemini Enterprise Agent Platform, Gemini 3.5 Flash, Omni, Antigravity 2.0, WebMCP, and the shift to generative AI search. The thread running through it all: agents are the headline, but governance and a solid semantic layer are the subplot that makes them actually useful.-----About The Measure Pod:The Measure Pod is your go-to fortnightly podcast hosted by seasoned analytics pros. Join Dara Fitzgerald (Co-Founder at Measurelab) & Matthew Hooson (Head of Engineering at Measurelab) as they dive into the world of data, analytics and measurement, with a side of fun.-----If you liked this episode, don't forget to subscribe to The Measure Pod on your favourite podcast platform and leave us a review. Let's make sense of the analytics industry together!

The entire startup ecosystem is racing to build agent harnesses. Logan Kilpatrick, who leads Google AI Studio and the Gemini API, argues that scramble has a roughly 12-month shelf life. Models will absorb the scaffolding and run it natively, so the edge moves elsewhere. Google's own bet runs in parallel: a single agent harness, born from the Windsurf team and now called Antigravity, has become the connective tissue across search, the Gemini app, Cloud, and AI Studio — the role Gemini-the-model used to play. Logan makes the case that coding already feels like narrow superintelligence, and that "jagged" vertical superintelligence (in math, finance, and science) will arrive well before AGI. He argues Google's real goal is maximizing outcomes for users, not eyeball time. He unpacks Omni, the single model built to replace multiple separate systems Google once trained for text, audio, music, image, and video. His throughline: AI is an accelerant for human ambition, not a substitute for it. Hosted by Sonya Huang, Sequoia Capital

The Nucleus LINK is used by more World Brewers Cup and World Barista Championship competitors than any other roasting machine. In this 65-minute webinar, Jerome Rosler and Sam Corra from Nucleus Coffee Tools walk through the exact roast profiles, competition case studies, and setup decisions that have taken athletes to the top of the podium — and answer your questions live.What you'll learn in this episode:— How the Brewers Cup score sheet works: aroma, flavor, aftertaste, acidity, sweetness, and mouthfeel — and which profile attributes move each score— Four real competition case studies with actual roast curves: Eileen, June, Justin, and Een (including a rare Eugenioides roasted at Singapore National Brewers Cup)— How to match the right Nucleus LINK profile to the cup attribute you want to maximize— The full story behind the Addis profile pack: from the 2021 World Championship foundations to Addis 4.1.2, the current release— How Jerome Rosler built the WBRC 2024 profile that helped Martin Wölfl win the World Brewers Cup— What Omni profiles are and how DTR lets you adapt a single coffee to any brew method— How to pack, seal, and travel with roasted competition coffee without killing your degassing curveWhat the experts say — key facts from this episode:Which Nucleus LINK profile works best for highly aromatic and fermented coffees?Filter E. Short total roast time (6:40–7 min), fast drying phase, development at 8.5–9% DTR over 30–40 seconds. Designed to highlight fruit-forward aromatics without amplifying heavy fermentation notes.How was Martin Wölfl's 2024 World Brewers Cup profile built?Based on filter C from the Addis 3.0 pack. Jerome Rosler shifted first crack 5 seconds earlier, raised the temperature increase rate from 6°/min to 7°/min, and shortened development to push solubility — the key variable for Wölfl's winning cup.What is the Nucleus LINK Addis profile pack?Addis is the current Nucleus LINK profile system. Named after the 2021 World Barista Championship host city, it has evolved across four versions: Addis 2.0 introduced filter profiles built around Nicole Battefeld-Montgomery's 2022 Worlds approach; Addis 3.0 refined filter C for Martin Wölfl's 2024 WBRC win; Addis 4.1.2 — the current version — adds espresso and Omni profiles for full brew-method coverage. Previous versions were also shaped by input from Sasa Sestic and Agnieszka Rojewska.How do Omni profiles work on the Nucleus LINK?Omni profiles use DTR (development time ratio) as the single adjustable variable, allowing the same coffee to be adapted across brew methods — from espresso to filter to cupping. Built to make washed and unconventional varieties competitive without needing separate profiles for each method.Where can you buy the Nucleus LINK in Europe and get profile support?Roast Rebels is Europe's authorized Nucleus Coffee Tools distributor and most experienced service center, with locations in Germany and Switzerland. Free shipping across the EU. Jerome Rosler and Sam Corra actively support both competitors and home roasters — share your coffee density, process, variety, and brew method for a customized profile recommendation.Links:Nucleus LINK Sample Roaster: roastrebels.com/en/nucleus-linkNucleus Coffee Tools: roastrebels.com/en/nucleus-coffee-toolsRoast Rebels Shop: roastrebels.com/enAbout Roast Rebels:Roast Rebels is Europe's go-to platform for specialty coffee roasting and Europe's most experienced Nucleus Coffee Tools distributor. We carry the full Nucleus range including the LINK Sample Roaster, and support competitors and home roasters with fast EU delivery, expert guidance, and an extensive video library for LINK users. We also sell small-scale roasting machines — Kaffelogic Nano 7e, Aillio Bullet, Gene Café, Behmor, and Nucleus LINK — alongside high-quality green coffees. Service centers in Germany and Switzerland.roastrebels.com/enroastrebels.com/en/nucleus-coffee-tools

The Power of Functional Medicine: Finding the Root Cause of Chronic Health Problems In this episode of Stay Healthy Knoxville, Dr. John-Mark Chesney sits down with Emily Turner, PA-C, and Randy Martin, PharmD, of Omni Functional Medicine to discuss a different approach to healthcare—one focused on identifying and addressing the root causes of chronic symptoms rather than simply managing them. Together, they explore what functional medicine is, why so many people continue to struggle despite being told their labs are normal, and how factors such as hormones, gut health, stress, metabolism, and inflammation can impact overall health and well-being. Whether you're dealing with fatigue, weight gain, digestive issues, autoimmune disease, chronic pain, or simply aren't feeling your best, this conversation offers valuable insight into how a more personalized approach to healthcare may help uncover missing pieces of the puzzle. In This Episode, You'll Learn: ✅ What functional medicine is and how it differs from traditional healthcare ✅ Why patients can still feel unwell despite "normal" lab results ✅ The importance of looking for root causes instead of just treating symptoms ✅ How hormones, gut health, stress, and metabolism influence overall health ✅ Practical steps you can take to improve your health and energy About Our Guests Emily Turner, PA-C Bachelor's Degree in Dietetics – University of Kentucky Master's in Physician Assistant Studies – Sullivan University Advanced training through The Institute of Functional Medicine Advanced training through The American Academy of Anti-Aging Medicine Randy Martin, PharmD Doctor of Pharmacy – University of Tennessee Co-founder of Omni Functional Medicine Connect with Omni Functional Medicine Instagram: @omnifunctionalmedicine Website: omnifunctionalmedicine.com Enjoying the Podcast? Be sure to subscribe, leave a review, and share this episode with someone who may be searching for answers to ongoing health concerns. Stay healthy, Knoxville!

In Part 1 of this two-part podcast series, Rebecca and Vickie share their firsthand experience touring a Disney Cruise ship and the Omni Convention Center in Florida. They discuss the ship's accommodations, dining options, entertainment, and family-friendly features, along with the convention center's event spaces and amenities. Their insights provide valuable information for travelers considering Disney cruises, group travel, conferences, and special events. Be sure to tune in for Part 2, where they continue sharing highlights and takeaways from their Florida travel industry experience. Follow us on all our social media accounts on Facebook and on Twitter at @Mousecapadespod. Thinking about being a guest on our show, or have a question or comment? Contact us anytime via text or phone at 636-373-4497. Have a magical day my friends!

Economics of AGI episode w Alex Imas and Phil Trammell.There's a bunch of important questions about how we deal with AI that only economics can answer.What is the optimal way to tax and redistribute the wealth that will be generated? How should countries not in the AI supply chain index into the gains? Is there any world where inequality doesn't explode?It might seem like these questions have obvious answers, but the first thing economics teaches you is that your intuitions can often be entirely wrong.It was very helpful to chat through these things with Alex and Phil.Watch on YouTube; read the transcript.SponsorsJane Street invests heavily in turning smart people into exceptional researchers and engineers. In addition to their apprenticeship model, Jane Street runs lectures and bootcamps in their in-office classrooms -- managers clear their teams' schedules to encourage attendance. If you'd like to work at a place that takes learning this seriously, Jane Street is hiring. Check out their open roles at janestreet.com/dwarkeshGoogle's Gemini Omni has incredible video editing capabilities -- you can upload a video and have Omni change the background, adjust lighting, or add specific elements. But Omni is also a preview of how future frontier models will be trained -- fully multimodal on both input and output. You can try it yourself in the Gemini app at gemini.google or in Flow at flow.googleCursor used targeted RL with textual feedback to help train their Composer 2.5 model. One of their researchers, Sasha Rush, gave me an impromptu blackboard lecture to explain how this form of on-policy self-distillation works -- I posted the full thing on X. If you want to try Composer 2.5, go to cursor.com/dwarkeshTimestamps(00:00:00) – Will capital share increase?(00:19:36) – Messy Middle scenario(00:25:57) – How to tax and redistribute AI wealth(00:30:02) – Why demand collapse is unlikely(00:39:26) – Human employees would be hard to integrate into the machine economy(00:43:08) – What if some humans (or AIs) value wealth accumulation intrinsically?(01:01:28) – What should developing countries do? Get full access to Dwarkesh Podcast at www.dwarkesh.com/subscribe

God is Omni

Kris Zellner is joined by Rob Naylor and Our Good Buddy Charles as we discuss the month of May 1991 in the world of World Championship Wrestling and pop culture at large. Topics of discussion include:The WWF trying to get dates at The Omni at a time when WCW was having some major issues drawing crowds at house shows.The wrestlers having a meeting about being overworked, hoping that their schedule will ease up soon.Rickey Henderson breaking the MLB stolen base record while Nolan Ryan throws his 7th no-hitter on the same day.TV season finales featuring “Night Court,” “A Different World,” “Beverly Hills 90210,” “Full House,” “In Living Color,” and the series finale of “Dallas.”The complete greatness of the video hyping up the Steiners vs. Lex Luger & Sting at SuperBrawl.Terry Funk appears on a wrestling themed episode of “Quantum Leap.”Reports of Hiroshi Hase & Kensuke Sasaki coming in…with manager Big Daddy Dink?!?!?Madonna's "Truth or Dare" and Bris Bosworth's “Stone Cold” hit the big screen.The TV ratings for WCW become dire, but they aren't alone in that.EMF, Seal, Smashing Pumpkins, and Jodeci all release their debut albums in the United States.President George Bush takes Queen Elizabeth to a baseball game.A full rundown of SuperBrawl, featuring the debuts of Johnny B. Badd, OZ, The Diamond Studd, and much more on a really fun PPV.This is just the tip of the iceberg, as we have so much going on during the month of May. I thought this was a tremendous show and I hope you agree!!!---To support the show and get access to exclusive rewards like special members-only monthly themed shows, go to our Patreon page at Patreon.com/BetweenTheSheets and become an ongoing Patron. Becoming a Between the Sheets Patron will also get you exclusive access to not only the monthly themed episode of Between the Sheets, but also access to our new mailbag segment, a Patron-only chat room on Slack, and anything else we do outside of the main shows!If you're looking for the best deal on a VPN service—short for Virtual Private Network, it helps you get around regional restrictions as well as browse the internet more securely—then Private Internet Access is what you've been looking for. Not only will using our link help support Between The Sheets, but you'll get a special discount, with prices as low as $1.98/month if you go with a 40 month subscription. With numerous great features and even a TV-specific Android app to make streaming easier, there is no better choice if you're looking to subscribe to WWE Network, AEW Plus, and other region-locked services.For the best in both current and classic indie wrestling streaming, make sure to check out IndependentWrestling.tv and use coupon code BTSPOD for a free 5 day trial! (You can also go directly to TinyURL.com/IWTVsheets to sign up that way.) If you convert to a paid subscriber, we get a kickback for referring you, allowing you to support both the show and the indie scene.To subscribe, you can find us on iTunes, Google Play, and just about every other podcast app's directory, or you can also paste Feeds.FeedBurner.com/BTSheets into your favorite podcast app using whatever “add feed manually” option it has.Advertising Inquiries: https://redcircle.com/brands

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,

[power press]

Day 13/100

[power press]

Arist Title Duration Album Year Composer Label Listeners Discipline Aria 10:11 Breadcrumbs (24-bit hi-res) 2025 38 1144 1:10 45 Karmamoi Aria 6:11 Odd Trip 2013 43 Karmakanic 1969 14:05 In a Perfect World 2011 Jonas Reingold/Julia Olsson 34 Karfagen The Glass of Time, Part 1 4:31 Omni II Act I: The Glass of Time 2026 37 Karfagen The Glass of Time, Part 1 4:31 Omni II Act I: The Glass of Time 2026 34 Karfagen Frozen Rivers 7:19 Omni II Act I: The Glass of Time 2026 35 The Flower Kings I am the Sun 5:33 Alive on Planet Earth Disc 1 2000 Century Media 36 Genesis Ripples [Live] 9:53 Genesis Archives, Vol. 2: 1976-1992 Disc 2 2000 Mike Rutherford 38 Karfagen Shadowbound 3:50 Omni II Act I: The Glass of Time 2026 37 Sunchild Father 10:21 Messages From Afar: The Division And Illusion Of Time 2018 39 1. Cry Boudica! 7:44 37 Karfagen The Shape of Love 8:32 Omni II Act I: The Glass of Time 2026 39 Sylvan Force of Gravity 5:10 Force of Gravity 2009 Sylvan Baukau 41 Karfagen Carry On 9:00 Omni II Act I: The Glass of Time 2026 38 Yogi Lang Move On 9:19 A Way Out Of Here 2019 35 RPWL Far Away From Home 4:27 Live From Outer Space 2019 36 Karfagen Beyond the Mirror 4:59 Omni II Act I: The Glass of Time 2026 36 Salva The Strong, Silent Type 8:18 A Thousand Ways To Disappear 2020 White Knight Records 35 Karfagen Omni (Part 6) (tracks 9 – 13) 19:17 Omni II Act I: The Glass of Time 2026 37

Kris Zellner is joined by Rob Naylor and Our Good Buddy Charles as we discuss the month of May 1991 in the world of World Championship Wrestling and pop culture at large. Topics of discussion include:The WWF trying to get dates at The Omni at a time when WCW was having some major issues drawing crowds at house shows.The wrestlers having a meeting about being overworked, hoping that their schedule will ease up soon.Rickey Henderson breaking the MLB stolen base record while Nolan Ryan throws his 7th no-hitter on the same day.TV season finales featuring “Night Court,” “A Different World,” “Beverly Hills 90210,” “Full House,” “In Living Color,” and the series finale of “Dallas.”The complete greatness of the video hyping up the Steiners vs. Lex Luger & Sting at SuperBrawl.Terry Funk appears on a wrestling themed episode of “Quantum Leap.”Reports of Hiroshi Hase & Kensuke Sasaki coming in…with manager Big Daddy Dink?!?!?Madonna's "Truth or Dare" and Bris Bosworth's “Stone Cold” hit the big screen.The TV ratings for WCW become dire, but they aren't alone in that.EMF, Seal, Smashing Pumpkins, and Jodeci all release their debut albums in the United States.President George Bush takes Queen Elizabeth to a baseball game.A full rundown of SuperBrawl, featuring the debuts of Johnny B. Badd, OZ, The Diamond Studd, and much more on a really fun PPV.This is just the tip of the iceberg, as we have so much going on during the month of May. I thought this was a tremendous show and I hope you agree!!!---To support the show and get access to exclusive rewards like special members-only monthly themed shows, go to our Patreon page at Patreon.com/BetweenTheSheets and become an ongoing Patron. Becoming a Between the Sheets Patron will also get you exclusive access to not only the monthly themed episode of Between the Sheets, but also access to our new mailbag segment, a Patron-only chat room on Slack, and anything else we do outside of the main shows!If you're looking for the best deal on a VPN service—short for Virtual Private Network, it helps you get around regional restrictions as well as browse the internet more securely—then Private Internet Access is what you've been looking for. Not only will using our link help support Between The Sheets, but you'll get a special discount, with prices as low as $1.98/month if you go with a 40 month subscription. With numerous great features and even a TV-specific Android app to make streaming easier, there is no better choice if you're looking to subscribe to WWE Network, AEW Plus, and other region-locked services.For the best in both current and classic indie wrestling streaming, make sure to check out IndependentWrestling.tv and use coupon code BTSPOD for a free 5 day trial! (You can also go directly to TinyURL.com/IWTVsheets to sign up that way.) If you convert to a paid subscriber, we get a kickback for referring you, allowing you to support both the show and the indie scene.To subscribe, you can find us on iTunes, Google Play, and just about every other podcast app's directory, or you can also paste Feeds.FeedBurner.com/BTSheets into your favorite podcast app using whatever “add feed manually” option it has.Advertising Inquiries: https://redcircle.com/brands

Google acaba de presentar uno de los eventos MÁS importantes de los últimos años… y sinceramente creo que mucha gente todavía no entendió lo que realmente pasó. Gemini Omni, agentes inteligentes, Android con IA total, nuevas gafas, generación de contenido, automatización extrema, búsquedas que podrían reemplazar páginas web… y una transformación gigantesca que afecta directamente a creadores, empresas, trabajos y al futuro entero de Internet. En este episodio analizamos TODO lo presentado en el Google I/O 2026, pero desde una mirada profunda, crítica y realista. Porque sí… lo que mostró Google es impresionante. Pero también abre preguntas enormes sobre privacidad, monetización, el futuro del trabajo y el rol de la inteligencia artificial en nuestras vidas. ¿Estamos entrando en una nueva era tecnológica… o en el comienzo de una crisis silenciosa para millones de personas? 

As the NCAA Division 1 men's golf participants descend on Omni La Costa in Carlsbad, CA, they can expect harder conditions with firmer and faster greens and grown up rough and fescue than previous years. Head Golf Professional of Resort and Tournament Golf, Robert Gogulich joins the Break80 Podcast to talk tournament setup and resort golf. He will discuss what he saw from the women's championship last week and what to expect when the men tee it up this week. Subscribe to the Break80 Podcast on Apple, Spotify and YouTube for weekly golf content. Learn more about your ad choices. Visit megaphone.fm/adchoices

What starts as Eric sharing a random dream immediately spirals into a full-scale debate about flying cars, teleportation, government tracking, magical transportation systems, and whether society can be trusted with a third dimension.Spoiler alert: probably not.In this episode:Eric dreams about driving a flying Dodge Omni inspired by Flight of the Navigator.A pickup baseball game in Martins Ferry somehow launches a 45-minute future-tech discussion.Why flying cars sound awesome until you remember other people exist.The horrifying logistics of airborne traffic jams, sky intersections, and people throwing soda cans from 2,000 feet.Disney technology, Spider-Man web shooters, and why Todd does not trust humanity with hover vehicles.The rise of teleportation as the superior transportation option.Teleportation ethics, naked arrival logistics, and the terrifying future of retina-scan Starbucks.How one Dodge Omni dream accidentally creates a dystopian sci-fi universe.From repulsor lift technology and magical sling rings to conspiracy theories about why we “aren't trying hard enough,” this episode somehow transforms into one of the most detailed conversations ever inspired by a car that should absolutely not be airborne.If you've ever wondered whether society is emotionally prepared for flying cars, the answer is probably no. But that won't stop Eric from trying to invent one.

The bracket continues with one of the strangest matchups yet: a disastrous piece of Star Wars history versus a flying Dodge Omni dream that somehow evolves into a full discussion about teleportation ethics.One episode is a structured roast of one of the worst television specials ever created. The other is two guys accidentally inventing a dystopian sci-fi future while talking about baseball pants.In this episode:A revisit to Spoiling Star Wars: The Holiday Special and why the special somehow gets worse with age.Abby's introduction to one of the most confusing pieces of Star Wars media ever created.Wookiee conversations, uncomfortable variety show energy, and the ongoing trauma of Life Day.A breakdown of The Flight of the Dodge Omni and how one random dream became a thirty-minute improv session.Flying cars, teleportation test pilots, and why Todd fundamentally does not trust humanity.The horrifying logistics of naked teleportation Starbucks hubs.A comparison between structured review episodes versus completely freeform conversational chaos.Honest discussion about audio quality, old recording setups, and why three microphones never sounded the same.What starts as a normal bracket recap slowly becomes another example of what the TodCast PodCast actually is at its core: two people taking ridiculous ideas far more seriously than necessary.One episode warns you not to watch something. The other warns you not to invent something. Only one moves on.

In Episode 84, our host Captain Ricky Wheeler talks with Martin Tollefsen of Simrad by Konsberg and Greg Mayer of the "Fishin' Frenzy" out of Oregon Inlet, NC. The topic of this podcast is all things Simrad Omni Sonar from the SY 50 to their new SY60. We also dive into what the new AI system from Viam does for Simrad Omni users. It's amazing how far technology has come and how this new AI for the Simrad Omni is making it easier for users to utilize this amazing tool. Martin dives into the finite details of the units, and Greg has a ton of first-hand user experience and input from his time using the SY50 on the water with a lot of success.To learn more about this amazing AI Sonar advancement CLICK HERETo reach Martin, you can email him at martin.tollefsen@simrad.com https://www.kongsberg.com/To reach out to Greg Mayer for a fishing trip on "Fishin' Frenzy" email him at greg@fishinfrenzy.comhttps://www.fishinfrenzy.com/If you would like our host, Ricky Wheeler, to help you sell your boat/yacht or help you with searching for and buying a boat/yacht, or booking a fishing charter you won't forget, please contact him at: https://www.saltwatereuphoria.com/contactSaltwater Euphoria Podcast Sponsors:+Saltwater Euphoria - https://www.saltwatereuphoria.com/+Billfish Gear - https://billfishgear.com/+Cape Maritime Consulting - https://www.capemaritimeconsulting.com/For online fishing courses, go to our website Courses.SaltwaterEuphoria.comFollow the following on Instagram:CaptainRickyWheeler: @CaptainRickyWheelerSaltwater Euphoria: @SaltwaterEuphoriaBillfish Gear: @billfishgearIf you like this podcast, please be sure to click that FOLLOW button and also spread the word by sharing this episode with your friends or whatever social channels you are on and/or leaving a great review.  We appreciate your support.

Joey returns from Google I/O with hands-on tests of Omni, Google's new video world model, comparing it head-to-head with Runway Aleph 2 on the same shots. Plus: Demis Hassabis puts AGI three years out, Google Flow gets agentic workflows, and Google Pic...

You'll need a map, compass and legend to understand all the new AI Google announced at its I/O conference last week. (They literally wrote a blog post called, "100 things we announced at I/O 2026” and most of them were AI based.) Luckily for you, we spend hours each day going through the latest in AI to cut the fluff from the real. So on today's ‘AI Working Wednesdays' series, we break down 3 of Google's biggest AI updates you can use today: Google Omni, Gemini 3.5 Flash and Antigravity 2.0. What's new and how do they work? We'll show you the ins and outs live. Newsletter: Sign up for our free daily newsletterMore on this Episode: Episode PageToday's Episode on LinkedIn: Thoughts on this? Join the convo on LinkedIn and connect with other AI leaders.Upcoming Episodes: Check out the upcoming Everyday AI Livestream lineupWebsite: YourEverydayAI.comEmail The Show: info@youreverydayai.comConnect with Jordan on LinkedInTopics Covered in This Episode:Gemini 3.5 Flash Model Hands-On DemoGemini 3.5 Flash Pricing and Token UsageBenchmarks: Gemini 3.5 Flash vs. 3.1 ProIntelligence vs. Cost in Gemini 3.5 FlashGemini 3.5 Flash for API and DevelopersGoogle Gemini Omni Flash Video Model ReviewOmni Anything-to-Anything Multimodal FeaturesGoogle Omni vs. Video Model CompetitorsAnti Gravity 2.0 Agent Desktop App OverviewAnti Gravity 2.0 Pros, Cons, and Use CasesUsage Limits in Google Gemini and Anti GravityChain of Thought Transparency in Gemini ModelsCanvas Mode Interactive Web App DemonstrationsTimestamps:00:00 Key AI updates from Google IO04:58 New Google AI updates discussed08:57 Google's anti gravity desktop use10:01 Touring Google's Anti Gravity App14:40 Testing a new AI prompt18:06 Critiquing vibe coding aesthetics21:28 Discussing Google's Gemini 3.1 Pro Model24:40 Comparing AI model performances and costs29:13 Google's advancements in video AI30:13 Future of Google's AI Technology33:58 Exploring Google Gemini features36:51 Google Gemini chain of thought feature42:02 Google Gemini's new model features44:23 River crossing puzzle gameplay48:25 Discussing Google Gemini 3.5 flash drawbacks51:10 Feedback on an AI releaseKeywords: Gemini 3.5 Flash, Google Gemini, AI updates, Google I/O 2026, Gemini Omni, Gemini Omni Flash, anti gravity 2.0, AI video model, hands-on AI demo, agentic coding, desktop AI app, benchmarking, AI model comparison, Gemini Spark, Gemini Pro 3.5, Gemini 3.1 Pro, token usage, API users, Google Workspace, always-on agent, AI cost efficiency, intelligent agents, world model, multimodal AI, generative video creation, video editing, scheduled tasks, Google Daily Brief, model usage limits, thinking steps, chain of thought, artificial analysis intelligence index, token inefficiency, cost to run AI, OpenAI GPT-5.5, Claude Sonnet, Claude Opus, open source AI models, AI-powered creativity, robotics, embodied AI, front-end AI tools, Canvas mode, conversational editing, interactive website builder, AI-powered app creation.Send Everyday AI and Jordan a text message. (We can't reply back unless you leave contact info) Start Here ▶️Not sure where to start when it comes to AI? Start with our Start Here Series. You can listen to the first drop -- Episode 691 -- or get free access to our Inner Cricle community and all episodes: StartHereSeries.com Also, here's a link to the entire series on a Spotify playlist. 

In "From Strategy to Scale: The ODW Logistics Approach to Growth" Joe Lynch and Phil Schmidbauer, Vice President of Solution Design at ODW Logistics, discuss how middle-market brands can scale by optimizing their entire supply chain network rather than just chasing low freight rates. About Phil Schmidbauer Phil Schmidbauer is the Vice President of Solution Design at ODW Logistics, where he specializes in creating optimized transportation and integrated supply chain strategies. A dynamic and innovative leader, Phil brings extensive industry experience focused on driving process efficiencies, eliminating waste, and delivering significant value to clients. Recognized as a top industry innovator—including being named a "Pros to Know" award winner—he excels at building strategic bridges across complex supply chain networks. Phil works closely with businesses to align their comprehensive logistics frameworks with overarching financial and operational goals. His expertise spans advanced supply chain analytics, cargo security, and network optimization studies, making him a trusted authority in helping growth-minded brands design custom-engineered solutions that reduce complexity and successfully position their businesses to scale. About ODW Logistics ODW Logistics is a top-tier, integrated third-party logistics (3PL) provider dedicated to enabling collective growth for its clients, associates, and the industry. With over 50 years of experience, ODW Logistics delivers end-to-end supply chain solutions that combine strategic warehousing, distribution, and advanced transportation management. The company serves a diverse range of industries, including food and beverage, consumer packaged goods, health and beauty, and industrial manufacturing. As an approved consolidator for major retail networks, ODW specializes in retail consolidation, strategic inventory load planning, and automated workflows that control costs and improve on-time delivery. Driven by core values of respect, trust, team, and opportunity, ODW Logistics operates as a seamless extension of its customers' businesses, providing the technology, infrastructure, and continuous innovation necessary to scale operations effectively. Key Takeaways: From Strategy to Scale: The ODW Logistics Approach to Growth In "From Strategy to Scale: The ODW Logistics Approach to Growth" Joe Lynch and Phil Schmidbauer, Vice President of Solution Design at ODW Logistics, discuss how middle-market brands can scale by optimizing their entire supply chain network rather than just chasing low freight rates. Integrated 3PL Solutions for Middle-Market Growth: ODW Logistics leverages over 50 years of experience to provide end-to-end warehousing, distribution, and managed transportation solutions, operating as a seamless extension for middle-market companies that lack the internal resources to manage complex supply chains alone. A Consultative, Total-Network Focus: Rather than just chasing the lowest transaction rate on a truck lane, Phil Schmidbauer emphasizes a consultative approach that designs and optimizes the entire supply chain, aligning warehousing and transportation around each other to reduce hidden costs, fines, and lead times. High-Frequency Retail Consolidation: ODW specializes in retail consolidation (serving major networks like Walmart and Target) by combining smaller multi-pallet shipments into full truckloads. This ensures high-frequency deliveries, which reduces lot sizes, minimizes inventory requirements, and drives better overall service. Mitigating the Cost of Stockouts: Keeping products on shelves is critical to brand survival. Stockouts cause severe financial penalties and permanent brand-loyalty loss when consumers switch to competitors—making consistent supply chain execution vital for sales growth. Managing the Hidden Costs of Excess Inventory: Influenced by his background with Toyota's world-class manufacturing processes, Schmidbauer highlights that excess inventory carries heavy hidden liabilities, including high warehousing fees, multiple touchpoints, and obsolescence or shelf-life expiration risks. The Power of a Dual-Node Network: ODW operates 27 facilities nationwide, utilizing a highly efficient dual-node setup between Southern California and Columbus, Ohio. This center-of-gravity strategy allows brands to easily meet next-day delivery demands for a massive portion of the U.S. population. Bridging the Omni-channel Divide: As retail and ecommerce models increasingly blend, ODW supports brands navigating both channels, helping companies scale and transition their operational structures from online-only to brick-and-mortar retail fulfillment seamlessly. Learn More About From Strategy to Scale: The ODW Logistics Approach to Growth Phil Schmidbauer | Linkedin ODW Logistics | Linkedin ODW Logistics The Logistics of Logistics Podcast If you enjoy the podcast, please leave a positive review, subscribe, and share it with your friends and colleagues. The Logistics of Logistics Podcast: Google, Apple, Castbox, Spotify, Stitcher, PlayerFM, Tunein, Podbean, Owltail, Libsyn, Overcast Check out The Logistics of Logistics on Youtube

Our 246th episode with a summary and discussion of last week's big AI news!Recorded on 05/22/2026Hosted by Andrey Kurenkov and Jeremie HarrisFeel free to email us your questions and feedback at andreyvkurenkov@gmail.com and/or hello@gladstone.aiRead out our text newsletter and comment on the podcast at https://lastweekin.ai/In this episode:Google I/O highlights included Gemini 3.5 (with 3.5 Flash emphasized for speed and benchmarks), the always-on agent Gemini Spark running on Google Cloud with MCP tool support, and Gemini Omni multimodal video generation/editing, plus updates like Anti-Gravity 2.0, Gemini for Science, and Genie world-model navigation using Street View and Waymo simulation.Coding-agent competition accelerated with Cursor Composer 2.5 (fine-tuned on Moonshot's Kimi K2.5) and xAI's early Grok Build release, alongside discussion of potential Cursor–xAI ties and xAI's talent churn and compute utilization concerns.Business and legal updates included Elon Musk losing his OpenAI lawsuit on statute-of-limitations grounds, reported OpenAI–Apple partnership tensions, Anthropic agreeing to a $30B funding round at a $900B valuation and projecting its first profitable quarter, and Cerebras' IPO surging about 90%. Research and safety stories covered OpenAI's result on an 80-year-old Erdős geometry problem, findings on “negation neglect” in training, interpretability work showing multiple redundant circuits per capability, agent benchmarks like Terminal World, new deepfake takedown enforcement under the Take It Down Act, demonstrations of autonomous hacking/self-replication, rapidly improving AI cyber capabilities, and steps toward image provenance metadata and watermarks.Timestamps:(00:00:10) Intro / Banter(00:01:15) News PreviewTools & Apps(00:05:05) Google unveils AI model Gemini 3.5 and AI agent Gemini Spark(00:11:43) Google's Gemini Omni turns images, audio, and text into video — and that's just the start | TechCrunch(00:17:27) Google launches Antigravity 2.0 with an updated desktop app and CLI tool at IO 2026 | TechCrunch(00:22:35) Google Debuts AI-Powered Tools To Optimize Scientific Research Workflows(00:27:20) Google's Genie world model can now simulate real streets with Street View | TechCrunch(00:29:51) Cursor's Composer 2.5 matches Opus 4.7 and GPT-5.5 benchmarks at a fraction of the cost(00:37:37) xAI Introduces Its Coding Agent Called Grok BuildApplications & Business(00:41:55) Musk loses OpenAI court battle as he waited too long to sue(00:48:08) Anthropic agrees terms of $30bn funding deal at $900bn valuation(00:53:12) OpenAI co-founder Andrej Karpathy joins Anthropic's pre-training team | TechCrunch(00:56:49) Greg Brockman Officially Takes Control of OpenAI's Products in Latest Shake-Up | WIRED(00:58:15) OpenAI-Apple Partnership Frays, Setting Up Possible Legal Fight - Bloomberg(01:01:13) AI chipmaker Cerebras soars 90% in year's biggest IPO so farResearch & Advancements(01:07:10) AI just solved an 80-year-old ‘Erdős problem,' and mathematicians are amazed | Scientific American(01:11:50) Negation Neglect: When models fail to learn negations in training(01:13:18) All Circuits Lead to Rome: Rethinking Functional Anisotropy in Circuit and Sheaf Discovery for LLMs(01:16:20) Autonomous AI research for nanogpt speedrun(01:21:59) TerminalWorld: Benchmarking Agents on Real-World Terminal TasksPolicy & Safety(01:23:15) America's dangerous, messy deepfakes crackdown is here | The Verge(01:25:17) Language Models Can Autonomously Hack and Self-Replicate(01:28:48) How fast is autonomous AI cyber capability advancing?(01:31:32) Positive Alignment: Artificial Intelligence for Human FlourishingSynthetic Media & Art(01:33:15) OpenAI is making it easier to check if an image was made by their models | TechCrunch(01:33:56) How Chinese short dramas became AI content machines | MIT Technology ReviewSee Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.

Drew and Rory are back for episode 69, which is legally required to begin with at least one immature joke before immediately collapsing under the weight of Google's latest AI product avalanche.This week, they dig into Google Omni, Gemini 3.5 Flash, Google Flow, Google Pics, Nano Banana, Veo, and whatever else Google launched before anyone had time to make coffee. The big question: are these actually meaningful creative upgrades, or did Google just throw 19 AI names into a blender and call it innovation?They break down early Omni and Flow tests, why video physics still feel weird, where Seedance and Kling may still be ahead, and why Runway Aleph 2.0 feels promising but imperfect. Rory shares hands-on examples with character swaps, driving videos, golf swings, agent mode, and Flow's new tool-building features. Drew tries to keep the conversation coherent while quietly wondering if every AI product now needs a map, glossary, and mild sedative.The episode also gets into Gemini as a search replacement, creepy context awareness, privacy tradeoffs, AI tools connecting to personal data, the fuzzy definition of “agentic,” the limits of auto-clipping tools, GPT Image 2's SynthID watermarking, metadata headaches for client work, and the universal pain of wasting $15 trying to make an image model spell “stump.”If you're trying to understand what Google's AI updates actually mean for creators, marketers, AI video workflows, image generation, creative direction, and the future of agentic media tools, this episode is half useful breakdown, half group therapy for people with too many tabs open.---⏱️ Fast Hour00:00 Cold open00:32 Google's AI naming avalanche01:39 AI hype vs actual workflow value02:34 Why AI launches feel like iPhone upgrades06:12 Google's “throw everything” strategy07:08 Omni vs Veo 4 expectations07:43 Video physics and speed problems09:03 Google Pics, Flow, Omni, and Flash10:04 How Rory actually uses Gemini11:51 Gemini 3.5 Flash breakdown12:38 AI benchmarks feel like marketing13:42 Gemini as a better search layer15:18 Creepy Gemini context awareness17:35 Why AI data connections feel too early19:15 The privacy tradeoff gets darker21:19 Google Omni vs Runway Aleph 2.022:12 Google Omni testing starts rough23:39 Google Veo 3.1 feels forgettable25:21 Why Omni feels early26:19 Higgsfield clipper test fails27:59 Why auto-clipping still misses31:30 Rory tests Flow and Omni live32:41 Omni character swap struggles33:33 Runway Aleph panda test34:07 Flow's new interface and tools35:02 Building custom tools inside Flow36:10 The joy of making tools from nothing37:39 Agent mode for still-image workflows39:05 Batch creative directions in Flow40:03 Omni turns six images into video40:47 Driving physics still feel off41:55 Why consistency matters for adoption43:03 Kling, Seedance, and the update race43:59 Seedance handles complex camera motion45:42 GPT Image setup for golf video46:53 Testing the same prompt in Flow49:25 Why agentic platforms can feel thin51:10 The need for visual design systems52:21 Flow's golf swing result53:56 Everyone is racing toward agentic54:18 What “agentic” actually means56:03 Claude feels more genuinely agentic57:04 Josh Hart quote analysis detour58:44 Reverse-engineering creative patterns59:53 Pizza, calzones, and prompt structure01:00:26 SynthID and GPT Image 2 watermarking01:01:47 Metadata problems for client work01:02:51 Google Pics enters the chat01:04:03 Too many image models to track01:04:52 Midjourney color still hits different01:06:01 GPT Image 2 quality frustration01:06:59 Image models still struggle with scale01:08:26 Bad AI weeks happen too01:09:20 Midjourney 8.2 speculation01:10:01 Tell your florist

Klik je týždenný komentovaný prehľad technologických správ, o udalostiach, ktoré sa udiali vo svete IT, médií a sociálnych sietí. Moderátori: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Ondrej Podstupka⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠, Martin Hodás Discord diskusný server nájdete tu: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://discord.gg/dAUW4PCaEh Linky: Google IO Spark https://www.techmeme.com/260519/p39#a260519p39 Omni https://www.techmeme.com/260519/p40#a260519p40 Vyhľadávanie https://www.techmeme.com/260519/p37#a260519p37 Karpathy https://www.axios.com/2026/05/19/anthropic-openai-karpathy-andrej-claude QR platby https://www.sme.sk/index/c/online-platby-na-mileticke-zaplatit-za-ceresne-trva-dve-minuty-terminaly-su-vynimkou Clean Room v Košiciach https://www.sav.sk/?lang=sk&doc=services-news&source_no=20&news_no=13630 Space Talk playlist https://www.youtube.com/playlist?list=PLNAJsgS6RlziDSV4GNWq8UXmSk81ETubA Spravili sme chybu, máte pripomienku? Napíšte nám na ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠klik@sme.sk⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ Kapitoly 00:00 Úvod01:14 Google AI novinky26:18 Karpathy v Anthropicu36:25 Microsoft Surface novinky41:00 Bezhotovostné platby pokračovanie45:50 Trump phone únik dát47:46 Starship štartuje52:04 ZáverSee omnystudio.com/listener for privacy information.

Google présente des agents IA capables d'agir à notre place • Un café suédois entièrement géré par une IA • Une puce quantique mille fois plus puissante • Tesla FSD autorisé en Lituanie • La voiture électrique accélère… mais les infrastructures suivront-elles ? • Les hôpitaux renforcent leur cybersécurité après des attaques massives.⭐️ Découvrez Frogans, l'innovation française qui réinvente le Web [PARTENARIAT]===============Google veut réinventer la recherche avec l'IA agentiqueLors de sa conférence annuelle, Google a dévoilé une transformation majeure de son moteur de recherche, désormais propulsé par Gemini 3.5 et orienté vers des usages “agentiques”. L'utilisateur ne se contentera plus d'obtenir des réponses : il pourra déléguer des tâches complexes, comparer, réserver, surveiller des informations ou générer des outils personnalisés. Ces annonces, détaillées dans un épisode spécial sur Monde Numérique, font craindre un bouleversement profond de l'écosystème du Web et du modèle économique des médias.Spark et Omni : l'IA personnelle et créative selon GoogleAvec Gemini Spark, Google promet un assistant capable d'interagir avec nos documents personnels et d'automatiser des flux de travail entiers. Côté création, Gemini Omni franchit un cap en permettant de modifier des vidéos existantes, d'y intégrer de nouveaux éléments ou de générer des scènes complètes à partir de contenus réels. Ces avancées ouvrent des perspectives inédites pour les créateurs… mais posent aussi des questions juridiques et économiques majeures.Elon Musk débouté face à OpenAIAux États-Unis, la justice a rejeté la plainte d'Elon Musk contre OpenAI pour des raisons procédurales. Le patron de Tesla accusait l'entreprise d'avoir trahi sa mission initiale à but non lucratif. Si la décision clôt provisoirement le volet judiciaire, elle ravive le débat sur l'évolution du modèle économique d'OpenAI et ses relations avec Microsoft.ChapsVision choisi par le renseignement allemandCocorico : le service de renseignement intérieur allemand a retenu la société française ChapsVision et sa plateforme ArgonOS pour moderniser ses capacités d'analyse de données. Un revers pour l'américain Palantir et un signal fort en faveur d'une souveraineté technologique européenne accrue dans les domaines sensibles.Tesla FSD autorisé en LituanieAprès les Pays-Bas, la Lituanie autorise à son tour le déploiement du FSD supervisé de Tesla. Le conducteur doit rester vigilant, mais cette étape marque une avancée supplémentaire vers l'autonomie en Europe. La France, elle, temporise encore malgré des tests réalisés à Paris.En Suède, une IA ouvre un café… et fait n'importe quoiÀ Stockholm, un café baptisé London Café est géré par une IA nommée Mona, basée sur Gemini. Budget, recrutement, commandes : tout est piloté par l'agent autonome. Résultat : des erreurs de gestion en cascade, des achats incohérents et un déficit important. L'expérience met en lumière les limites actuelles des modèles en matière de mémoire et de cohérence opérationnelle à long terme.Une puce japonaise aux performances révolutionnairesDes chercheurs de l'Université de Tokyo ont présenté dans la revue Science un composant exploitant la spintronique et la commutation quantique. La promesse : des calculs mille fois plus rapides et une consommation divisée par cent, avec une dissipation thermique minimale. Si l'industrialisation reste à venir, cette avancée pourrait transformer l'efficacité énergétique des data centers.Bruno Guglielminetti, Mon Carnet, lance un flash tech en 10 langues grâce à l'IADepuis Montréal, Bruno Guglielminetti analyse les annonces de Google et leurs implications pour les créateurs de contenus. Il présente également son nouveau flash “120 secondes de tech”, désormais disponible en dix langues grâce à un système d'agents IA développé avec la société Productivia, démontrant concrètement l'automatisation avancée de la production éditoriale.Recharge ultra-rapide : la voiture électrique change d'échelle[PARTENARIAT] Alors que les constructeurs chinois annoncent des vitesse de recharge de véhicules électriques ultra rapides, Julien Villeret, directeur de l'innovation d'EDF, détaille les défis liés aux puissances de charge, au refroidissement et à l'adaptation des réseaux électriques, dans un contexte géopolitique tendu et de transition accélérée vers l'électrique.Cybersécurité hospitalière : retour d'expérience après une attaque majeure[PARTENARIAT] À l'occasion du salon SantExpo, en partenariat avec la Fédération hospitalière de France, Nasser Amani, directeur des services numériques des hôpitaux Nord-Ouest, revient sur la cyberattaque subie en 2021. Il décrit l'arrêt brutal des systèmes, la gestion en mode dégradé et les leçons tirées pour renforcer la résilience des établissements face à des centaines de milliers de tentatives d'intrusion mensuelles.Hébergé par Audiomeans. Visitez audiomeans.fr/politique-de-confidentialite pour plus d'informations.

¿Ha cumplido Google con las expectativas o nos han vendido humo?

Анонсы nFactorial, рекомендации из рассылки nFactorial Weekly, переход Андрея Карпаты в Anthropic, итоги Google I/O разбор возможностей видеомодели Gemini Omni, анализ интервью Джеффа Безоса о капитализме, налогах и бизнес-стратегии, стэнфордское исследование об опасности чрезмерной вежливости ИИ и подстраивании под пользователя, инвестиции Сэма Альтмана в стартапы Y Combinator через токены OpenAI, сборник фундаментальных советов для стартапов от Y Combinator, соревнование по сортировке посылок между человеком и гуманоидным роботом Figure, график инфляции за 25 лет и дефляционная природа технологий на фоне роста цен на услуги, научное исследование влияния 20-секундных объятий на уровень стресса и окситоцина, вирусные видео с главой Nvidia Дженсеном Хуангом, создание персонального ИИ-агента главой МИД Сингапура, разбор устройства AlphaGo в подкасте Дваркеша Пателя, оценка главных бенефициаров потенциального IPO компании SpaceX, смена карьерных приоритетов, почему роль High Impact Individual Contributor стала престижнее руководящих должностей, а также истории успеха из Instagram nFactorial. Рекомендации от nFactorial - Создаем команду Agentic AI-инженеров. Подробнее: https://www.linkedin.com/feed/update/urn:li:activity:7463104249846034433/ - nFactorial AI Cup: открытый чемпионат Казахстана по вайб-кодингу веб-игр (24 мая, 9:00-18:00, Нархоз Университет, призовой фонд - 1.1 млн тенге + 3 гранта на nFactorial Incubator) - https://www.instagram.com/p/DYE_O6MjW_x/ - nFactorial Reunion - Встреча выпускников nFactorial Incubator разных лет: где они сейчас, что делали тогда, советы - https://www.youtube.com/watch?v=Csr4j8vAcco - Подписаться на nFactorial Weekly - https://nfactorial-school.kit.com/

Google I/O showed a company with enormous AI advantages and a surprisingly confusing product map. NLW breaks down Omni, Spark, Antigravity 2.0, Gemini 3.5 Flash, and the deeper strategic question underneath it all: whether Google is really trying to beat Claude Code and Codex at their own game, or whether its real bet is on consumer distribution, multimodal world models, TPUs, and embedding AI across everything people already use.Apply for our Growth Engineering role: ⁠⁠⁠⁠⁠⁠https://jobs.aidailybrief.ai/⁠⁠⁠⁠⁠⁠Enterprise Claw Cohort 3 Registration: ⁠⁠⁠⁠https://enterpriseclaw.ai/⁠⁠⁠⁠Brought to you by:KPMG – Agentic AI is powering a potential $3 trillion productivity shift, and KPMG's new paper, Agentic AI Untangled, gives leaders a clear framework to decide whether to build, buy, or borrow—download it at ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠www.kpmg.us/Navigate⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Granola - The AI notepad for people in back-to-back meetings. 100% off your first 3 months with code AIDAILY at ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠http://granola.ai/aidaily⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Scrunch - The AI customer experience platform - ⁠⁠⁠⁠⁠⁠https://scrunch.com/⁠⁠⁠⁠⁠⁠Mercury - Modern banking for business and now personal accounts. Learn more at ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://mercury.com/personal-banking⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Zenflow Work - Agents for knowledge work - ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://zenflow.free/⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Drata - The agentic trust management platform - ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://drata.com/⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Blitzy - Want to accelerate enterprise software development velocity by 5x? ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://blitzy.com/⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠AssemblyAI - The best way to build Voice AI apps - ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.assemblyai.com/brief⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Robots & Pencils - Cloud-native AI solutions that power results ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://robotsandpencils.com/⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠The AI Daily Brief helps you understand the most important news and discussions in AI. Subscribe to the podcast version of The AI Daily Brief wherever you listen: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://pod.link/1680633614⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Our Newsletter is BACK: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://aidailybrief.beehiiv.com/⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Interested in sponsoring the show? sponsors@aidailybrief.ai

Google I/O 2026 just dropped Gemini Omni, a world-model AI that simulates physics, edits video, and might be the biggest leap since Seedance 2. But it's not perfect. Gavin and Kevin break down everything from Google I/O 2026, including the launch of Gemini Omni (Google's new world model), Gemini 3.5 Flash benchmarks against GPT-5.5 and Opus 4.7, the Gemini Spark personal agent, AskYouTube, Docs Live, new AI glasses, the first search box redesign in 25 years, and the shocking news that Andrej Karpathy is joining Anthropic. SHOW LINKS: Google I/O 2026 Full Keynote: https://www.youtube.com/live/wYSncx9zLIU?si=Nb881MfGTlf1Q0II Gemini Omni physics demos from Google DeepMind: https://x.com/GoogleDeepMind/status/2056786449312493669?s=20 Gemini Omni's incredible London knowledge (via fofrAI): https://x.com/fofrAI/status/2056789242274259242?s=20 Sundar Pichai and Demis Hassabis on Omni video editing: https://x.com/sundarpichai/status/2056524502746747048?s=20 Gavin's hands-on Gemini Omni experiments: https://x.com/gavinpurcell/status/2056762427879182692?s=20 Gemini Omni's character cameo feature (less impressive): https://x.com/gavinpurcell/status/2056772793539481830?s=20 Gemini Omni volleyball fail: https://x.com/flavioAd/status/2056771223359549645?s=20 Google's new Content Credentials Verification: https://x.com/Google/status/2056787498676658576?s=20 Genie 3 IRL — Google's world model now simulates real streets with Street View: https://techcrunch.com/2026/05/19/googles-genie-world-model-can-now-simulate-real-streets-with-street-view/ Bilawal Sidhu on Genie 3 IRL: https://x.com/bilawalsidhu/status/2056804315721843024?s=20 Gemini 3.5 Flash launches — official announcement: https://x.com/GeminiApp/status/2056788115893993701?s=20 Gemini Spark — Google's new personal coding agent: https://x.com/Google/status/2056791134295273554?s=20 Google's new AI glasses  https://x.com/backlon/status/2056807059707036050?s=20 Andrej Karpathy joins Anthropic to focus on recursive self-learning: https://www.axios.com/2026/05/19/anthropic-openai-karpathy-andrej-claude  

Logan Kilpatrick and Tulsee Doshi of Google DeepMind join for a first-ever in-person episode recorded just days before Google I/O, covering headline launches like Gemini 3.5 Flash, the Omni video generation model, and the new Gemini Spark agentic product. The conversation digs into Google's strategic decision to lead with cost-adjusted efficiency over raw capability, how DeepMind now ships a full agent harness rather than bare models, and technical questions around context window limits and knowledge cutoffs. They also explore how the team thinks about model psychology, AI welfare, and recursive self-improvement. Sponsors: Brave Search API: Brave Search API gives AI agents a fast, independent search index for research, RAG pipelines, images, places, and fewer hallucinations. Get $5 in free credits at https://brave.com/search/api/?mtm_campaign=q2-26-cognitive-revolution Sequence: Sequence handles the full revenue workflow for complex pricing, from quoting and metering to invoicing, revenue recognition, and collections. Book a public demo at https://sequencehq.com and use code COGNISM in the source field to save 20% off year one Roboflow: Roboflow is an end-to-end visual AI platform that lets you turn raw ideas into fully deployed applications in just hours, powering breakthroughs like Blueprint Pro's floor-plan understanding tool. Read the full Blueprint Pro story and see how over a million engineers are building the next wave of visual AI at https://roboflow.com Claude: Claude by Anthropic is an AI collaborator that understands your workflow and helps you tackle research, writing, coding, and organization with deep context. Get started with Claude and explore Claude Pro at https://claude.ai/tcr

Host Matt Paige records a special Talking AI episode live from Google I/O with AI creators Kushank Aggarwal, Marcin Teodoru, and Jay Enrique, discussing Google's biggest announcements and what will matter in real use.They argue Google's edge is distribution—bringing AI to existing Search users—positioning Gemini as an intelligence layer across products like Search, YouTube, Gmail, Docs, Chrome, Android, and shopping.They highlight rapid growth in token usage, Search's new AI mode and generative UI/dashboard experiences, and YouTube features that jump to relevant video moments, potentially improving discoverability for creators and local businesses.They debate Gemini Spark's agentic approach, prepackaged agents like Daily Brief, and enterprise “agent garden” concepts, then cover Omni as a broader “world model” play, Pix/NanoBanana-style editing and image workflow improvements, and a glasses demo featuring translation, Gemini Live, and impressive audio.--Key Moments:00:54 Gemini Everywhere Strategy02:09 Search Gets Agentic03:47 Generative UIs for All06:48 YouTube as Action Engine08:21 Gemini Spark Agents10:10 Adoption and Standards13:55 Omni World Model17:13 Pix Editing Workflow19:06 Omni Platform Take19:53 Fire Round Highlights22:05 Glasses Demo Reactions24:02 Wrap Up and Where to Follow--Key Links:DigitalSamaritanConnect with Kushank on LinkedInRoboNuggetsConnect with Jay on LinkedInAI BuildersConnect with Marcin on LinkedIn

Are you creating content every single day and still feeling invisible? The problem isn't how much you're posting, it's your diversification strategy. In this episode of The Visibility Impact Show, visibility strategist Crissy Conner breaks down the exact marketing ecosystem you need to be visible across social media, search, and AI in 2026.Crissy introduces the four pillars of a strong visibility strategy:A long-term marketing strategy (YouTube, podcasting, or blogging)A short-term social media strategyA keyword and search strategy so the right people can find youA leverage strategy that builds credibility over timeFind out how to work with Crissy at https://thevisibleceo.com/workwithcrissyOMNI is my full visibility system built for CEOs who want to grow online without living on their phone. If you're ready to be truly seen, more strategic, and unmistakably in demand, head to check out OMNI at www.omniqueens.com https://www.instagram.com/itscrissyconner/https://www.tiktok.com/@crissyconnerhttps://www.facebook.com/crissyconnerhttps://www.youtube.com/c/crissyconnerhttps://www.linkedin.com/in/crissyconner/

Jeannette is joined by Dr. Guy Sandelowsky, co-founder of the award-winning plant-powered pet food brand, Omni. Guy reflects on the business's phenomenal journey over the past 12 months, detailing how their appearance on Dragons' Den acted as a massive catalyst to 10x their top-line revenue.  From surviving a brutal grilling by Peter Jones to securing backing from Steven Bartlett and Deborah Meaden, he shares invaluable insights into scaling a startup, navigating corporate sustainability through B Corp status, and the unique challenges faced as an LGBTQ+ founder in the business world. You'll Learn Why: Appearing on Dragons' Den can serve as a powerful catalyst for unprecedented business growth by providing massive, free national television exposure. Sustained growth relies on establishing a clear product-market fit by solving a genuine, widespread pain point for your consumers. Achieving B Corp certification is not just about ethical practice; it creates a highly resilient business model that aligns with the values of modern consumers and drives higher revenue growth. Navigating investor bias by remaining fiercely authentic to your identity can ultimately unlock the right opportunities and secure partners who truly support your vision. This episode is living proof that no matter where you're starting from — or what life throws at you — it's never too late to be brave, bold, and unlock your inner brilliant. Visit https://brave-bold-brilliant.com/ for free tools, guides and resources to help you take action now

In episode 68 of Fast Hours, Drew and Rory return from a two-week hiatus to prove that yes, the AI news cycle did continue without their permission. Rude.They dig into Freepik changing its name to Magnific, why enterprise AI image tools are starting to feel more like creative operating systems, and how brands may be better off using approved model aggregators instead of building weird internal Franken-tools that immediately become outdated.Then things get nerdier. Obviously.Rory breaks down how he's using Codex, GPT-Image-2, Claude Code, MCPs, Higgsfield, Seedance, and visual style reference sheets to create repeatable image systems, character references, and bulk creative workflows without living inside a giant text prompt forever. Drew pushes into where Midjourney V8.1 still dominates, especially photorealistic faces, color, texture, and images that do not look like corporate stock photography that lost the will to live.They also talk about Midjourney's upcoming 8.2, 8.3, V9 roadmap, edit model ambiguity, personalization drift, Luma Uni comparisons, Pinterest's internal AI image model, Salesforce going headless, and why AI video audio still sounds like it was recorded inside a cursed podcast booth.And because no episode is complete without accidentally getting philosophical, they close with the viral Claude Monet AI social experiment, the weird bias people bring to AI-generated images, and why “how it was made” keeps hijacking whether people can actually see what's in front of them.Basically, it's an episode about the future of AI creative tools, with two guys trying to sound calm while the ground turns into soup beneath them.--⏱️ Fast Hour00:00 Cold open01:12 AI news fatigue is real01:44 Claude Code runs the day now03:11 Remote work and coffee shop crimes08:21 3 Ninjas nostalgia break10:09 Freepik becomes Magnific11:47 Why Magnific works for enterprise13:00 Model aggregators vs internal tools18:01 Pinterest builds its own AI image model22:11 Salesforce goes "headless"24:08 Higgsfield, MCPs, and Meta ads29:51 Codex for GPT-Image-2 workflows32:45 Pulling style from video frames34:05 Building visual style reference sheets37:36 Codex and textured illustration systems39:34 The evolution beyond text prompts41:50 Seedance storyboards and visual prompts43:01 Reference images as reusable seeds44:51 AI video still has an audio problem46:13 Audio reference hacks in Dreamina50:52 Omni-reference for video control52:37 Midjourney V8.1 updated take53:28 The blue and pink problem returns55:37 Midjourney still owns realistic faces58:56 Reworking old prompts with Describe01:01:10 Luma Uni vs Midjourney color01:02:11 Midjourney 8.2, 8.3, and V901:03:46 Midjourney edit model questions01:07:13 Midjourney plus Seedance films01:09:03 Midjourney's strange lane01:12:28 The Monet AI social experiment01:15:13 Why people over-detect AI01:20:09 AI backlash and disclosure debates01:22:50 AI as a career unlock01:24:30 Keep making weird stuff01:25:45 Wrap-up and seamstress CTA

Thanks to  @HPInc  & Intel for sponsoring us! More on the Zbook Fury https://bit.ly/4uapNHs Google I/O is next week and the AI leaks are pouring out: a new Spark agent, Veo 4 Omni, Gemini 3.2 Flash that's reportedly 20x cheaper than GPT-5.5. This week on AI For Humans, Google is cooking again and the I/O leaks are stacking up. We dig into Google Spark, a new Gemini agent that may have access to your entire digital life. Veo 4 Omni model leaks suggest deeper reasoning and character consistency, and the model gets math right. Gemini 3.2 Flash is rumored to deliver 90% of GPT-5.5's capability at a fraction of the cost and dramatically faster speeds. There's a new GoogleBook with Gemini built in. And Google is reinventing the mouse cursor, the input device that's been largely unchanged since 1968, with voice AI. Plus, Thinking Machines dropped voice interactivity demos that feel a lot like ChatGPT Voice from two years ago. OpenAI is reportedly already working on GPT-5.6, and Sam Altman is giving away two free months of Codex to companies to drive adoption. Gavin's been experimenting with local open-source LLMs and shares his setup.  AND…we get into the data center sickness conversation: infrasound from data centers may be causing cortisol spikes in nearby communities. Figure 03's package sorting livestream proved the robot is autonomous after skeptics accused it of being teleoperated. Unitree dropped a transformable robot.  AI KEEPING US UP AT NIGHT. NO MATTER. WE COOK. // Show Links // Google Spark: Gemini's Agent With Access To Your Life https://x.com/kimmonismus/status/2054855742247584231?s=20 Veo 4 Omni Model Leaks: Gets Math Right https://x.com/TomLikesRobots/status/2053845600051798065?s=20 More Veo 4 Omni Examples https://x.com/testingcatalog/status/2053718756799467735?s=20 Omni Model Added To Gemini Web Build https://x.com/testingcatalog/status/2054196983523393857?s=20 Gemini 3.2 Flash At 90% Of GPT-5.5 For Way Less https://x.com/kimmonismus/status/2054887891222802633?s=20 New GoogleBook With Gemini Built In https://x.com/Google/status/2054270454467121187?s=20 Google DeepMind: Rethinking The Mouse Cursor With Voice AI https://deepmind.google/blog/ai-pointer Thinking Machines Voice Interactivity Demos https://thinkingmachines.ai/blog/interaction-models/ Sam Altman: Two Months Of Free Codex For Companies https://x.com/sama/status/2054626219858293128?s=20 Data Center Sickness: Ben Jordan's Video On Infrasound https://youtu.be/_bP80DEAbuo Figure 03 Package Sorting Livestream https://www.youtube.com/live/luU57hMhkak?si=KZHwUdYUwY4SIRUp Brett Adcock: Figure 03 Was Not Teleoperated https://x.com/adcock_brett/status/2054737974710169840?s=20 Unitree Transformable Robot https://x.com/UnitreeRobotics/status/2054067819634159622?s=20