Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

Diffusion is the foundational ML framework behind state-of-the-art AI image and video generation, including Sora, Midjourney and Google Veo. In this episode of Decoded, Ankit Gupta sits down with Francois Chaubard to discuss how diffusion works, walk through a code sample, and explain why everyone training models should understand... show more

Y Combinator

1,539,544 subscribers

74,604 views • 6 months ago •via X (Twitter)

Education Science & Technology

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

I'm thrilled to announce the launch of ⚡️Flash Diffusion from Jasper! Earlier this year, with our acquisition of Clipdrop, we launched the Jasper AI Research Lab in Paris. Today, we are excited to release our first piece of groundbreaking research: the open-source distillation method, "Flash Diffusion". Flash Diffusion accelerates inference by 500%, reduces computing costs, and produces higher-quality image outputs. Dive into the details and discover how Flash Diffusion is set to revolutionize the field of AI and image synthesis. Read all about it here: Try a demo on Hugging Face:

I'm thrilled to announce the launch of ⚡️Flash Diffusion from Jasper! Earlier this year, with our acquisition of Clipdrop, we launched the Jasper AI Research Lab in Paris. Today, we are excited to release our first piece of groundbreaking research: the open-source distillation method, "Flash Diffusion". Flash Diffusion accelerates inference by 500%, reduces computing costs, and produces higher-quality image outputs. Dive into the details and discover how Flash Diffusion is set to revolutionize the field of AI and image synthesis. Read all about it here: Try a demo on Hugging Face:

Timothy Young

10,093 views • 2 years ago

Google isn’t betting on a single AI architecture. Sundar Pichai, CEO of Google: “We’re going to push the diffusion paradigm as hard as possible.” “All of today’s mainline Gemini models are autoregressive. Diffusion is a different paradigm.” “For the same capability, diffusion can be much faster.” “It’s behind the mainline models today, but there will be areas where it’s the right tool.” “We’re pushing multiple directions in parallel, and bringing them together where it makes sense.”

Google isn’t betting on a single AI architecture. Sundar Pichai, CEO of Google: “We’re going to push the diffusion paradigm as hard as possible.” “All of today’s mainline Gemini models are autoregressive. Diffusion is a different paradigm.” “For the same capability, diffusion can be much faster.” “It’s behind the mainline models today, but there will be areas where it’s the right tool.” “We’re pushing multiple directions in parallel, and bringing them together where it makes sense.”

Forward Future

152,896 views • 6 months ago

Full interview with the founder and CEO of Propr Lou sits down with me to discuss Hyperliquid and the future of prop trading Timestamps: 00:00:00 Intro 00:00:16 Meet Louis: from TradFi to crypto 00:01:52 How the crypto market has changed 00:03:49 Why build on Hyperliquid 00:06:52 How propr got started 00:08:40 Scaling a lean team (and the grind) 00:12:24 What a prop firm actually is 00:15:33 Agents, bots, and the future of trading 00:18:32 What Louis is most proud of 00:21:02 Thinking about the token and TGE 00:24:14 Where propr goes in 5 years 00:25:57 Best and worst trades 00:27:50 Life outside crypto + shoutouts

Full interview with the founder and CEO of Propr Lou sits down with me to discuss Hyperliquid and the future of prop trading Timestamps: 00:00:00 Intro 00:00:16 Meet Louis: from TradFi to crypto 00:01:52 How the crypto market has changed 00:03:49 Why build on Hyperliquid 00:06:52 How propr got started 00:08:40 Scaling a lean team (and the grind) 00:12:24 What a prop firm actually is 00:15:33 Agents, bots, and the future of trading 00:18:32 What Louis is most proud of 00:21:02 Thinking about the token and TGE 00:24:14 Where propr goes in 5 years 00:25:57 Best and worst trades 00:27:50 Life outside crypto + shoutouts

ryandcrypto

42,534 views • 1 month ago

Why do even our best AI models need tens of thousands of examples to learn skills that a human picks up in a handful of tries? Solving this problem is one of the great open challenges in modern AI. World models, which give AI an internal simulation of its environment, are one of the most promising paths forward. In this episode of Decoded, YC's Ankit Gupta and Francois Chaubard discuss the intuition and math behind world models, new research, and current applications in self-driving, robotics, and more. 01:45 — What would perfect efficiency look like? 05:10 — World models in the human brain 09:20 — Control theory & the drone example 14:30 — When physics breaks down 17:45 — Chess, Go & the action space problem 24:10 — Why AlphaGo can't scale 28:00 — Monte Carlo tree search explained 34:00 — Self-Driving: state space is infinite 40:30 — Model-Free vs. Model-Based RL 44:00 — Why robotics is the hardest case 48:20 — World models that actually work 54:10 — JEPA & latent space tricks 59:00 — Open problems remaining 1:04:30 — Does this pass the squint test?

Why do even our best AI models need tens of thousands of examples to learn skills that a human picks up in a handful of tries? Solving this problem is one of the great open challenges in modern AI. World models, which give AI an internal simulation of its environment, are one of the most promising paths forward. In this episode of Decoded, YC's Ankit Gupta and Francois Chaubard discuss the intuition and math behind world models, new research, and current applications in self-driving, robotics, and more. 01:45 — What would perfect efficiency look like? 05:10 — World models in the human brain 09:20 — Control theory & the drone example 14:30 — When physics breaks down 17:45 — Chess, Go & the action space problem 24:10 — Why AlphaGo can't scale 28:00 — Monte Carlo tree search explained 34:00 — Self-Driving: state space is infinite 40:30 — Model-Free vs. Model-Based RL 44:00 — Why robotics is the hardest case 48:20 — World models that actually work 54:10 — JEPA & latent space tricks 59:00 — Open problems remaining 1:04:30 — Does this pass the squint test?

Y Combinator

108,267 views • 5 days ago

My conversation with OpenAI co-founder Greg Brockman This is the most detailed first-person account of the 72 hours after Sam Altman was fired. We also go deep on what comes next: the global race to AGI, why ChatGPT stopped showing reasoning, how much of OpenAI's own code is now written by AI ("it's hard to know what percent is not"), and the untold story of how OpenAI actually started in 2015. 00:00:00 Introduction 00:00:49 Meeting Sam Altman and Starting OpenAI 00:02:40 Building the Founding Team 00:04:25 DeepMind's Lead Over OpenAI 00:04:54 Changing OpenAI to a For-Profit Model 00:06:05 Breakthrough Moments at OpenAI 00:08:22 What Dota 2 Meant for OpenAI 00:10:04 Reasoning Versus Prediction 00:11:59 Tensions Grow at OpenAI 00:15:44 Sam Altman's Firing 00:17:49 Greg Quits OpenAI 00:19:56 Sam Explores Deal with Microsoft's Satya 00:20:28 Petition for Altman's Return 00:23:43 Ilya Sutskever Leaves OpenAI 00:24:59 Lessons Learned after Sam Ousting 00:28:22 The Thing Ilya Said that Greg Can't Forget 00:32:22 Is AI Going Parabolic? 00:33:24 How Much of OpenAI's Code is Written by AI? 00:36:21 Do AI Chatbots Tell Us What We Want to Hear? 00:38:06 The Global AI Race to Reach AGI 00:38:40 What Happens if US Doesn't Reach AGI First? 00:39:49 Are Countries Stealing AI Advancements? 00:40:38 Why ChatGPT No Longer Shows Reasoning 00:41:47 The Finite Constraints of Compute 00:43:38 On Investing Early in Data Centers 00:46:31 The Future of Data Center Specialization 00:47:52 How to Decide Whose Queries to Serve 00:49:08 OpenAI on Consumer vs Enterprise Models 00:53:05 Data Centers in Space? 01:00:56 What Should AI Regulation Look Like? 01:04:33 The Future of AI-Powered Entrepreneurship 01:04:44 AI and Job Loss 01:07:15 The Skills Young People Should Invest In 01:11:30 What Does Success Look Like For You? Full episode on X below. Also find it on: • YouTube: • Spotify: • Apple:

My conversation with OpenAI co-founder Greg Brockman This is the most detailed first-person account of the 72 hours after Sam Altman was fired. We also go deep on what comes next: the global race to AGI, why ChatGPT stopped showing reasoning, how much of OpenAI's own code is now written by AI ("it's hard to know what percent is not"), and the untold story of how OpenAI actually started in 2015. 00:00:00 Introduction 00:00:49 Meeting Sam Altman and Starting OpenAI 00:02:40 Building the Founding Team 00:04:25 DeepMind's Lead Over OpenAI 00:04:54 Changing OpenAI to a For-Profit Model 00:06:05 Breakthrough Moments at OpenAI 00:08:22 What Dota 2 Meant for OpenAI 00:10:04 Reasoning Versus Prediction 00:11:59 Tensions Grow at OpenAI 00:15:44 Sam Altman's Firing 00:17:49 Greg Quits OpenAI 00:19:56 Sam Explores Deal with Microsoft's Satya 00:20:28 Petition for Altman's Return 00:23:43 Ilya Sutskever Leaves OpenAI 00:24:59 Lessons Learned after Sam Ousting 00:28:22 The Thing Ilya Said that Greg Can't Forget 00:32:22 Is AI Going Parabolic? 00:33:24 How Much of OpenAI's Code is Written by AI? 00:36:21 Do AI Chatbots Tell Us What We Want to Hear? 00:38:06 The Global AI Race to Reach AGI 00:38:40 What Happens if US Doesn't Reach AGI First? 00:39:49 Are Countries Stealing AI Advancements? 00:40:38 Why ChatGPT No Longer Shows Reasoning 00:41:47 The Finite Constraints of Compute 00:43:38 On Investing Early in Data Centers 00:46:31 The Future of Data Center Specialization 00:47:52 How to Decide Whose Queries to Serve 00:49:08 OpenAI on Consumer vs Enterprise Models 00:53:05 Data Centers in Space? 01:00:56 What Should AI Regulation Look Like? 01:04:33 The Future of AI-Powered Entrepreneurship 01:04:44 AI and Job Loss 01:07:15 The Skills Young People Should Invest In 01:11:30 What Does Success Look Like For You? Full episode on X below. Also find it on: • YouTube: • Spotify: • Apple:

Shane Parrish

450,952 views • 3 months ago

DreamCraft3D: Hierarchical 3D Generation with Bootstrapped Diffusion Prior paper page: present DreamCraft3D, a hierarchical 3D content generation method that produces high-fidelity and coherent 3D objects. We tackle the problem by leveraging a 2D reference image to guide the stages of geometry sculpting and texture boosting. A central focus of this work is to address the consistency issue that existing works encounter. To sculpt geometries that render coherently, we perform score distillation sampling via a view-dependent diffusion model. This 3D prior, alongside several training strategies, prioritizes the geometry consistency but compromises the texture fidelity. We further propose Bootstrapped Score Distillation to specifically boost the texture. We train a personalized diffusion model, Dreambooth, on the augmented renderings of the scene, imbuing it with 3D knowledge of the scene being optimized. The score distillation from this 3D-aware diffusion prior provides view-consistent guidance for the scene. Notably, through an alternating optimization of the diffusion prior and 3D scene representation, we achieve mutually reinforcing improvements: the optimized 3D scene aids in training the scene-specific diffusion model, which offers increasingly view-consistent guidance for 3D optimization. The optimization is thus bootstrapped and leads to substantial texture boosting. With tailored 3D priors throughout the hierarchical generation, DreamCraft3D generates coherent 3D objects with photorealistic renderings, advancing the state-of-the-art in 3D content generation.

DreamCraft3D: Hierarchical 3D Generation with Bootstrapped Diffusion Prior paper page: present DreamCraft3D, a hierarchical 3D content generation method that produces high-fidelity and coherent 3D objects. We tackle the problem by leveraging a 2D reference image to guide the stages of geometry sculpting and texture boosting. A central focus of this work is to address the consistency issue that existing works encounter. To sculpt geometries that render coherently, we perform score distillation sampling via a view-dependent diffusion model. This 3D prior, alongside several training strategies, prioritizes the geometry consistency but compromises the texture fidelity. We further propose Bootstrapped Score Distillation to specifically boost the texture. We train a personalized diffusion model, Dreambooth, on the augmented renderings of the scene, imbuing it with 3D knowledge of the scene being optimized. The score distillation from this 3D-aware diffusion prior provides view-consistent guidance for the scene. Notably, through an alternating optimization of the diffusion prior and 3D scene representation, we achieve mutually reinforcing improvements: the optimized 3D scene aids in training the scene-specific diffusion model, which offers increasingly view-consistent guidance for 3D optimization. The optimization is thus bootstrapped and leads to substantial texture boosting. With tailored 3D priors throughout the hierarchical generation, DreamCraft3D generates coherent 3D objects with photorealistic renderings, advancing the state-of-the-art in 3D content generation.

AK

161,530 views • 2 years ago

How does math research change when the cost of trying your first dumb idea goes to zero? Daniel Litt joins Greg Burnham and Anson Ho to discuss what today’s models can and can’t do in math, and how far they are from doing high-quality research. 0:00:00 What's the hardest math problem AI can solve today? 00:16:08 How helpful are today’s AI models for math research? 00:23:36 Junk papers, LLM-generated proofs, and the refereeing crisis 00:27:21 AI enables searching through problems at scale 00:33:49 When will AI be good enough to publish in top math journals? 00:42:15 What are the returns to intelligence? 00:59:50 Will AI solve Millennium problems? 01:11:54 Is math full of low-hanging fruit? 01:18:47 How Daniel has adapted his professional life to AI progress 01:25:28 What do AI math benchmarks actually measure? 01:33:05 Designing the Open Problems benchmark 01:56:35 Do mathematicians believe heuristic arguments about conjectures? 02:01:24 What if FrontierMath: Open Problems gets solved? 02:06:53 Is AI on the cusp of accelerating math progress?

How does math research change when the cost of trying your first dumb idea goes to zero? Daniel Litt joins Greg Burnham and Anson Ho to discuss what today’s models can and can’t do in math, and how far they are from doing high-quality research. 0:00:00 What's the hardest math problem AI can solve today? 00:16:08 How helpful are today’s AI models for math research? 00:23:36 Junk papers, LLM-generated proofs, and the refereeing crisis 00:27:21 AI enables searching through problems at scale 00:33:49 When will AI be good enough to publish in top math journals? 00:42:15 What are the returns to intelligence? 00:59:50 Will AI solve Millennium problems? 01:11:54 Is math full of low-hanging fruit? 01:18:47 How Daniel has adapted his professional life to AI progress 01:25:28 What do AI math benchmarks actually measure? 01:33:05 Designing the Open Problems benchmark 01:56:35 Do mathematicians believe heuristic arguments about conjectures? 02:01:24 What if FrontierMath: Open Problems gets solved? 02:06:53 Is AI on the cusp of accelerating math progress?

Epoch AI

178,703 views • 5 months ago

This is a big day. Meta is open-sourcing AudioCraft. You can now generate incredible music and sounds with a single prompt. It includes the most performant Generative AI Model (audio) on the market, the "Llama" of Audio. The research framework contains the weights and code of these models: ▸ MusicGen: controllable text-to-music model. ▸ AudioGen: text-to-sound model. ▸ EnCodec: high fidelity neural audio codec. ▸ Multi Band Diffusion: An EnCodec compatible decoder using diffusion. This is going to tremendously speed up audio research 👏

This is a big day. Meta is open-sourcing AudioCraft. You can now generate incredible music and sounds with a single prompt. It includes the most performant Generative AI Model (audio) on the market, the "Llama" of Audio. The research framework contains the weights and code of these models: ▸ MusicGen: controllable text-to-music model. ▸ AudioGen: text-to-sound model. ▸ EnCodec: high fidelity neural audio codec. ▸ Multi Band Diffusion: An EnCodec compatible decoder using diffusion. This is going to tremendously speed up audio research 👏

Lior Alexander

231,608 views • 3 years ago

Google DeepMind Developers: How Nano Banana Was Made In this episode, we sat down with Google DeepMind’s Oliver Wang and Nicole Brichtova to discuss how Nano Banana was created, why it became so viral, and what’s in store for the future of image and video editing. 00:00 Intro 02:00 The origin of Nano Banana 06:20 Seeing yourself in AI 11:00 Control & character consistency 17:10 AI in education and visual learning 24:10 2D vs 3D: Debate over world models 31:10 The Japan phenomenon 35:00 From images to video 41:00 Working with artists 47:30 The next era of image models 53:50 Closing thoughts Oliver Wang Nicole Brichtova Guido Appenzeller Yoko

Google DeepMind Developers: How Nano Banana Was Made In this episode, we sat down with Google DeepMind’s Oliver Wang and Nicole Brichtova to discuss how Nano Banana was created, why it became so viral, and what’s in store for the future of image and video editing. 00:00 Intro 02:00 The origin of Nano Banana 06:20 Seeing yourself in AI 11:00 Control & character consistency 17:10 AI in education and visual learning 24:10 2D vs 3D: Debate over world models 31:10 The Japan phenomenon 35:00 From images to video 41:00 Working with artists 47:30 The next era of image models 53:50 Closing thoughts Oliver Wang Nicole Brichtova Guido Appenzeller Yoko

a16z

97,860 views • 8 months ago

Today, we released Microsoft's newest AI Diffusion Report, and I sat down with Juan M. Lavista Ferres, Chief Data Scientist and Director of the AI for Good Lab, for a special edition of #ToolsAndWeapons to unpack the report’s most striking insights. It’s a conversation about technology’s reach, the pace of progress, and what it means for the world. The report reveals how generative AI is spreading globally—and why the speed of diffusion will determine who benefits most. We spotlight countries leading the charge like UAE, Singapore, and South Korea, and explore the rise of models such as China’s DeepSeek across parts of Africa. Listen to the full episode:

Today, we released Microsoft's newest AI Diffusion Report, and I sat down with Juan M. Lavista Ferres, Chief Data Scientist and Director of the AI for Good Lab, for a special edition of #ToolsAndWeapons to unpack the report’s most striking insights. It’s a conversation about technology’s reach, the pace of progress, and what it means for the world. The report reveals how generative AI is spreading globally—and why the speed of diffusion will determine who benefits most. We spotlight countries leading the charge like UAE, Singapore, and South Korea, and explore the rise of models such as China’s DeepSeek across parts of Africa. Listen to the full episode:

Brad Smith

85,263 views • 6 months ago

We don't know what most microbial genes do. Can genomic language models help? there's only one way to find out! this is a 1 hour and 42 minute interview with an MIT professor (the famous Yunha Hwang) chatting about these questions, her work in solving them at Tatta Bio, and more. zoomer captions are back too Links in reply! Timestamps: 00:00:00 - Clips + sponsor roll from the wonderful LatchBio 00:02:07 – Introduction 00:02:23 – Why do microbial genomes matter 00:04:07 – Deep learning acceptance in metagenomics 00:05:25 – The case for genomic “context” over sequence matching 00:06:43 – OMG: the only ML-ready metagenomic dataset 00:09:27 – gLM2: A multimodal genomic language model 00:11:06 – What do you do with the output of genomic language models? 00:17:41 – How will OMG evolve? 00:20:26 – Why train on only microbial genomes, as opposed to all genomes? 00:22:58 – Do we need more sequences or more annotations? 00:23:54 – Is there a conserved microbial genome ‘language’? 00:28:11 – What non-obvious things can this genomic language model tell you? 00:33:08 – Semantic deduplication and evaluation 00:37:33 – How does benchmarking work for these types of models? 00:41:31 – Gaia: A genomic search engine 00:44:18 – Even ‘well-studied’ genomes are mostly unannotated 00:50:51 – Using agents on Gaia 00:54:53 – Will genomic language models reshape the tree of life? 00:59:18 – Current limitations of genomic language models 01:08:54 – Directed evolution as training data 01:12:35 – What is Tatta Bio? 01:19:02 – Building Google for genomic sequences (SeqHub) 01:25:46 – How to create communities around scientific OSS 01:29:06 – What’s the purpose in the centralization of the software? 01:35:37 – How will the way science is done change in 10 years?

We don't know what most microbial genes do. Can genomic language models help? there's only one way to find out! this is a 1 hour and 42 minute interview with an MIT professor (the famous Yunha Hwang) chatting about these questions, her work in solving them at Tatta Bio, and more. zoomer captions are back too Links in reply! Timestamps: 00:00:00 - Clips + sponsor roll from the wonderful LatchBio 00:02:07 – Introduction 00:02:23 – Why do microbial genomes matter 00:04:07 – Deep learning acceptance in metagenomics 00:05:25 – The case for genomic “context” over sequence matching 00:06:43 – OMG: the only ML-ready metagenomic dataset 00:09:27 – gLM2: A multimodal genomic language model 00:11:06 – What do you do with the output of genomic language models? 00:17:41 – How will OMG evolve? 00:20:26 – Why train on only microbial genomes, as opposed to all genomes? 00:22:58 – Do we need more sequences or more annotations? 00:23:54 – Is there a conserved microbial genome ‘language’? 00:28:11 – What non-obvious things can this genomic language model tell you? 00:33:08 – Semantic deduplication and evaluation 00:37:33 – How does benchmarking work for these types of models? 00:41:31 – Gaia: A genomic search engine 00:44:18 – Even ‘well-studied’ genomes are mostly unannotated 00:50:51 – Using agents on Gaia 00:54:53 – Will genomic language models reshape the tree of life? 00:59:18 – Current limitations of genomic language models 01:08:54 – Directed evolution as training data 01:12:35 – What is Tatta Bio? 01:19:02 – Building Google for genomic sequences (SeqHub) 01:25:46 – How to create communities around scientific OSS 01:29:06 – What’s the purpose in the centralization of the software? 01:35:37 – How will the way science is done change in 10 years?

owl

44,279 views • 7 months ago

She literally broke down how to run evals in Claude Code (built the whole thing live): 01:34 - What people get wrong with evals 04:35 - Why product taste is the alpha now 09:28 - Building a PM agent from one prompt 19:00 - Instrumentation without writing code 22:00 - Watching traces stream in live 28:00 - Getting Claude to write your first eval 33:58 - When vibe evals work and when they don't 48:50 - The self-improving loop (this part is wild) 01:03:00 - Same-day shipping is real 01:06:00 - The context graph unlock

She literally broke down how to run evals in Claude Code (built the whole thing live): 01:34 - What people get wrong with evals 04:35 - Why product taste is the alpha now 09:28 - Building a PM agent from one prompt 19:00 - Instrumentation without writing code 22:00 - Watching traces stream in live 28:00 - Getting Claude to write your first eval 33:58 - When vibe evals work and when they don't 48:50 - The self-improving loop (this part is wild) 01:03:00 - Same-day shipping is real 01:06:00 - The context graph unlock

Aakash Gupta

150,362 views • 2 months ago

$A 7-million parameter model outperforming models a thousand times its size on tasks like ARC Prize. That's what recursive reasoning unlocks. In this episode of Decoded, YC's Ankit Gupta and Francois Chaubard break down two recent papers on recursive AI models, HRMs and TRMs, that are achieving state-of-the-art results with a fraction of the parameters of today's largest models. They explain why standard LLMs hit a fundamental ceiling on certain reasoning tasks, how recursion at inference time gives small models the compute depth to break through it, and what happens when you combine these ideas with the power of large-scale foundation models. 00:35 - Model Foundations 01:15 - RNN Limits and LLM Contrast 02:36 - Reasoning Limits and Sorting Analogy 04:22 - HRM Paper Introduction 05:25 - HRM Architecture and Intuition 07:36 - HRM Results and Outer Loop 09:46 - TRM Paper Overview 11:20 - TRM Training and Fixed Point 13:30 - Detailed HRM Summary 20:46 - Comparing HRM and TRM 34:45 - Future Outlook$

A 7-million parameter model outperforming models a thousand times its size on tasks like ARC Prize. That's what recursive reasoning unlocks. In this episode of Decoded, YC's Ankit Gupta and Francois Chaubard break down two recent papers on recursive AI models, HRMs and TRMs, that are achieving state-of-the-art results with a fraction of the parameters of today's largest models. They explain why standard LLMs hit a fundamental ceiling on certain reasoning tasks, how recursion at inference time gives small models the compute depth to break through it, and what happens when you combine these ideas with the power of large-scale foundation models. 00:35 - Model Foundations 01:15 - RNN Limits and LLM Contrast 02:36 - Reasoning Limits and Sorting Analogy 04:22 - HRM Paper Introduction 05:25 - HRM Architecture and Intuition 07:36 - HRM Results and Outer Loop 09:46 - TRM Paper Overview 11:20 - TRM Training and Fixed Point 13:30 - Detailed HRM Summary 20:46 - Comparing HRM and TRM 34:45 - Future Outlook

Y Combinator

127,645 views • 2 months ago

I'll remember this week for the rest of my life. Here's my interview with the incredible Sundar Pichai, CEO of Google. We spoke about Gemini, world models, self-improving AI, AI anxiety, and more. Full vid below: 0:00 - Intro 0:40 - Gemini as a World Model 1:30 - Gemini Diffusion 2:33 - Self-improving AI 4:09 - What AI needs to improve 5:06 - Agent Memory 6:46 - AI Form Factors 7:58 - Google Search Future 9:33 - AI Anxiety

I'll remember this week for the rest of my life. Here's my interview with the incredible Sundar Pichai, CEO of Google. We spoke about Gemini, world models, self-improving AI, AI anxiety, and more. Full vid below: 0:00 - Intro 0:40 - Gemini as a World Model 1:30 - Gemini Diffusion 2:33 - Self-improving AI 4:09 - What AI needs to improve 5:06 - Agent Memory 6:46 - AI Form Factors 7:58 - Google Search Future 9:33 - AI Anxiety

Matthew Berman

98,864 views • 1 year ago

New episode of the Information Bottleneck! We talked with Stefano Ermon about why he thinks diffusion LLMs will replace autoregressive ones. Stefano co-invented DDIM, FlashAttention, DPO, and score-based diffusion models. He's a Stanford professor and now runs Inception AI, where they built Mercury II. We go deep but also cover the bigger picture - the startup journey, PhD vs industry, and where AI is heading. A few things that stuck with me: - He thinks of autoregressive models as typewriters and diffusion models as editors. One goes left to right. The other starts messy and refines. - Mercury II (their text difussion model) already beats the fastest autoregressive models on latency-critical stuff as voice agents, code suggestions, anything where you have a tight time budget. And it does it because diffusion generates tokens in parallel instead of one at a time. - We also got into whether AI will actually replace software engineers (his answer: no), PhD vs industry advice, and what it was like going from an ICML best paper to raising money.

New episode of the Information Bottleneck! We talked with Stefano Ermon about why he thinks diffusion LLMs will replace autoregressive ones. Stefano co-invented DDIM, FlashAttention, DPO, and score-based diffusion models. He's a Stanford professor and now runs Inception AI, where they built Mercury II. We go deep but also cover the bigger picture - the startup journey, PhD vs industry, and where AI is heading. A few things that stuck with me: - He thinks of autoregressive models as typewriters and diffusion models as editors. One goes left to right. The other starts messy and refines. - Mercury II (their text difussion model) already beats the fastest autoregressive models on latency-critical stuff as voice agents, code suggestions, anything where you have a tight time budget. And it does it because diffusion generates tokens in parallel instead of one at a time. - We also got into whether AI will actually replace software engineers (his answer: no), PhD vs industry advice, and what it was like going from an ICML best paper to raising money.

Ravid Shwartz Ziv

22,350 views • 4 months ago

Aidan Gomez (Aidan Gomez) is a computer scientist, co-author of the seminal paper ‘Attention Is All You Need,’ and the CEO of Cohere. In this episode, we discussed his upbringing in the cabin his grandfather built in Codrington, a small town North of Brighton, and the values that were instilled in him through his family. We explored his path from Codrington to his undergraduate studies at the University of Toronto, emailing Geoffrey Hinton, and joining Google Brain where he co-wrote the paper on transformers. We discussed how he met his co-founders Ivan Zhang and Nick Frosst, and his insights on what it means to build a meaningful, successful company in Canada. Aidan shares his conviction about what is at stake — for Canada and for the world at large. This is a conversation about family, values, and what it means to live with conviction. The Other Stuff is hosted by internetVin — filmmaker, entrepreneur, and possibly the most curious man on Earth. Produced by New. The Other Stuff #29 — Aidan Gomez: Empathy and Conviction — Timestamps 00:00:00 Intro 00:03:10 The Malleability of Toronto 00:11:06 Growing Up in Codrington 00:13:50 The Story of Aidan Gomez’s Family 00:27:23 Introduction to the Internet 00:30:39 Values and Work Ethic 00:35:46 University of Toronto’s AI Scene 00:40:48 Emailing Geoffrey Hinton 00:42:16 Google Brain 00:45:15 Dropout: A Simple Way to Prevent Neural Networks from Overfitting 00:49:48 The Beauty of Research 00:54:24 One Model to Rule Them All 00:59:54 Meeting Ivan Zhang and Nick Frosst 01:04:00 The Birth of Cohere 01:06:56 Twitter Influencers and Alex Friedland 01:11:48 Being the CEO 01:12:51 Building for Canada 01:15:01 Three Fundamental Ingredients of Building a Company 01:21:39 Working with the Canadian Government 01:24:39 Reflexivity in Canada 01:36:22 What Is Evil? 01:42:40 The Role of AI in the World

Aidan Gomez (Aidan Gomez) is a computer scientist, co-author of the seminal paper ‘Attention Is All You Need,’ and the CEO of Cohere. In this episode, we discussed his upbringing in the cabin his grandfather built in Codrington, a small town North of Brighton, and the values that were instilled in him through his family. We explored his path from Codrington to his undergraduate studies at the University of Toronto, emailing Geoffrey Hinton, and joining Google Brain where he co-wrote the paper on transformers. We discussed how he met his co-founders Ivan Zhang and Nick Frosst, and his insights on what it means to build a meaningful, successful company in Canada. Aidan shares his conviction about what is at stake — for Canada and for the world at large. This is a conversation about family, values, and what it means to live with conviction. The Other Stuff is hosted by internetVin — filmmaker, entrepreneur, and possibly the most curious man on Earth. Produced by New. The Other Stuff #29 — Aidan Gomez: Empathy and Conviction — Timestamps 00:00:00 Intro 00:03:10 The Malleability of Toronto 00:11:06 Growing Up in Codrington 00:13:50 The Story of Aidan Gomez’s Family 00:27:23 Introduction to the Internet 00:30:39 Values and Work Ethic 00:35:46 University of Toronto’s AI Scene 00:40:48 Emailing Geoffrey Hinton 00:42:16 Google Brain 00:45:15 Dropout: A Simple Way to Prevent Neural Networks from Overfitting 00:49:48 The Beauty of Research 00:54:24 One Model to Rule Them All 00:59:54 Meeting Ivan Zhang and Nick Frosst 01:04:00 The Birth of Cohere 01:06:56 Twitter Influencers and Alex Friedland 01:11:48 Being the CEO 01:12:51 Building for Canada 01:15:01 Three Fundamental Ingredients of Building a Company 01:21:39 Working with the Canadian Government 01:24:39 Reflexivity in Canada 01:36:22 What Is Evil? 01:42:40 The Role of AI in the World

The Other Stuff Podcast

45,297 views • 7 months ago

We've got a podcast! In our first episode, Ege, Tamay and Jaime dig into: • What they expect AI to look like by 2030 • Why economists are underestimating the likelihood of explosive growth • The startling regularity in technological trends like Moore's Law • Moravec’s paradox, and how we might overcome it And more more! Timestamps: 00:00:00 Preview 00:00:37 What is Epoch AI? 00:02:32 Scaling Laws 00:08:43 Key Drivers 00:19:20 End of the Decade Predictions 00:21:18 Bottlenecks: Power 00:27:59 Bottlenecks: GPUs 00:32:07 Bottlenecks: Data 00:45:37 Bottlenecks: Latency 00:56:18 Bottlenecks: Failure Rates 01:03:55 AI Investment 01:07:11 Automation 01:12:10 Benchmarks & Moravec’s Paradox 01:19:45 Economic Impact 01:45:48 Open Questions & Takeaways

We've got a podcast! In our first episode, Ege, Tamay and Jaime dig into: • What they expect AI to look like by 2030 • Why economists are underestimating the likelihood of explosive growth • The startling regularity in technological trends like Moore's Law • Moravec’s paradox, and how we might overcome it And more more! Timestamps: 00:00:00 Preview 00:00:37 What is Epoch AI? 00:02:32 Scaling Laws 00:08:43 Key Drivers 00:19:20 End of the Decade Predictions 00:21:18 Bottlenecks: Power 00:27:59 Bottlenecks: GPUs 00:32:07 Bottlenecks: Data 00:45:37 Bottlenecks: Latency 00:56:18 Bottlenecks: Failure Rates 01:03:55 AI Investment 01:07:11 Automation 01:12:10 Benchmarks & Moravec’s Paradox 01:19:45 Economic Impact 01:45:48 Open Questions & Takeaways

Epoch AI

152,591 views • 1 year ago

Third episode of Nerd Snipe is live! This time I try to get Ben in on my Anthropic conspiracy theories (and explain why Claude got dumber) 00:07:30 T3 Code Banned? 00:16:08 How Claude Caching Works 00:24:00 Why Claude Seems Dumber 00:38:03 The Conspiracies Begin...

Third episode of Nerd Snipe is live! This time I try to get Ben in on my Anthropic conspiracy theories (and explain why Claude got dumber) 00:07:30 T3 Code Banned? 00:16:08 How Claude Caching Works 00:24:00 Why Claude Seems Dumber 00:38:03 The Conspiracies Begin...

Theo - t3.gg

202,397 views • 3 months ago

Upscale-A-Video: Temporal-Consistent Diffusion Model for Real-World Video Super-Resolution paper page: Text-based diffusion models have exhibited remarkable success in generation and editing, showing great promise for enhancing visual content with their generative prior. However, applying these models to video super-resolution remains challenging due to the high demands for output fidelity and temporal consistency, which is complicated by the inherent randomness in diffusion models. Our study introduces Upscale-A-Video, a text-guided latent diffusion framework for video upscaling. This framework ensures temporal coherence through two key mechanisms: locally, it integrates temporal layers into U-Net and VAE-Decoder, maintaining consistency within short sequences; globally, without training, a flow-guided recurrent latent propagation module is introduced to enhance overall video stability by propagating and fusing latent across the entire sequences. Thanks to the diffusion paradigm, our model also offers greater flexibility by allowing text prompts to guide texture creation and adjustable noise levels to balance restoration and generation, enabling a trade-off between fidelity and quality. Extensive experiments show that Upscale-A-Video surpasses existing methods in both synthetic and real-world benchmarks, as well as in AI-generated videos, showcasing impressive visual realism and temporal consistency.

Upscale-A-Video: Temporal-Consistent Diffusion Model for Real-World Video Super-Resolution paper page: Text-based diffusion models have exhibited remarkable success in generation and editing, showing great promise for enhancing visual content with their generative prior. However, applying these models to video super-resolution remains challenging due to the high demands for output fidelity and temporal consistency, which is complicated by the inherent randomness in diffusion models. Our study introduces Upscale-A-Video, a text-guided latent diffusion framework for video upscaling. This framework ensures temporal coherence through two key mechanisms: locally, it integrates temporal layers into U-Net and VAE-Decoder, maintaining consistency within short sequences; globally, without training, a flow-guided recurrent latent propagation module is introduced to enhance overall video stability by propagating and fusing latent across the entire sequences. Thanks to the diffusion paradigm, our model also offers greater flexibility by allowing text prompts to guide texture creation and adjustable noise levels to balance restoration and generation, enabling a trade-off between fidelity and quality. Extensive experiments show that Upscale-A-Video surpasses existing methods in both synthetic and real-world benchmarks, as well as in AI-generated videos, showcasing impressive visual realism and temporal consistency.

AK

32,849 views • 2 years ago