正在加载视频...

视频加载失败

返回首页

🚀 AI Deep Dive Episode 3: Recursive Self-Improvement AI can design better chips. Better chips can build better AI. 🚀 Rocket and Ricursive Intelligence co-founders Anna Goldie and Azalia Mirhoseini go deep on the hardware behind recursive self-improvement—and what happens when that loop starts accelerating. – Intro – What...

theinformation's profile picture

The Information

1,598,911 subscribers

17,374 次观看 • 1 个月前 •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

3 条评论

The Information 的头像
The Information29 天前

00:00 – Intro 01:18 – What Is Recursive Self-Improvement? 03:18 – Can AI Automate Chip Design? 11:10 – Co-Designing AI and Hardware 19:35 – Beyond Today’s Models and Chips 26:08 – How Much More Efficient Could AI Get? 31:31 – What Happens When the Loop Accelerates? 42:14 – Can Alignment Keep Up? 47:05 – Should We Slow AI Progress?

مستر حكيم 的头像
مستر حكيم1 个月前

في حلقة "AI Deep Dive Episode 3"، نستكشف مفهومًا مثيرًا للاهتمام: تحسين الذات التكراري للذكاء الاصطناعي. إن فكرة أن الذكاء الاصطناعي يمكنه تصميم رقائق أكثر كفاءة، والتي بدورها تُمكّن من بناء ذكاء اصطناعي أقوى، هي حلقة مفرغة إيجابية قد تقود إلى تطورات هائلة. تسلط المؤسستان، @rocketalignment و@RicursiveAI، الضوء على أهمية الأجهزة المتخصصة في تسريع عمليات التعلم والتحسين الذاتي للذكاء الاصطناعي. إنها خطوة أساسية نحو تحقيق مستوى أعلى من الذكاء الاصطناعي العام، حيث يمكن للنظام تحسين نفسه بشكل مستمر، مما يزيد من قدراته المعرفية بمرور الوقت. تُظهر هذه الحلقة كيف أن التقدم في مجال الأجهزة يمكن أن يكون له تأثير مضاعف على تطور الذكاء الاصطناعي، مما يدفعنا إلى آفاق جديدة من الإمكانيات والابتكار.

赵战阳 的头像
赵战阳29 天前

@rocketalignment @RicursiveAI @grok 总结全文

相关视频

Introducing Ricursive Intelligence, a frontier AI lab enabling a recursive self-improvement loop between AI and the chips that fuel it. Learn more at
4:04

Introducing Ricursive Intelligence, a frontier AI lab enabling a recursive self-improvement loop between AI and the chips that fuel it. Learn more at

Ricursive Intelligence

534,090 次观看 • 11 个月前

Chip design takes years. At RAISE Summit, Ricursive says AI can dramatically speed up the process—and that better chips will in turn enable better AI models, creating a recursive cycle of improvement.
1:24

Chip design takes years. At RAISE Summit, Ricursive says AI can dramatically speed up the process—and that better chips will in turn enable better AI models, creating a recursive cycle of improvement.

TIME

7,214,348 次观看 • 3 个月前

“If we can make AI make better chips, then those chips can power the next generation of AI models.” Startup ‘Ricursive Intelligence,’ founded by Azalia Mirhoseini and Anna Goldie, has been building AI tools to do just that: to speed up the chip design process. The company’s name is a nod to “recursive self-improvement,” the idea that AI could one day bootstrap itself without human involvement. If they’re successful, Goldie and Mirhoseini could revolutionize the way the chip industry works. At the moment, AI companies mostly optimize their models for certain chips. But in a future where chip design happens far faster, it could work the other way around: chips optimized for specific models. The founders are on the TIME100 AI, the list of people shaping the development of AI in 2026. See the full list:
1:12

“If we can make AI make better chips, then those chips can power the next generation of AI models.” Startup ‘Ricursive Intelligence,’ founded by Azalia Mirhoseini and Anna Goldie, has been building AI tools to do just that: to speed up the chip design process. The company’s name is a nod to “recursive self-improvement,” the idea that AI could one day bootstrap itself without human involvement. If they’re successful, Goldie and Mirhoseini could revolutionize the way the chip industry works. At the moment, AI companies mostly optimize their models for certain chips. But in a future where chip design happens far faster, it could work the other way around: chips optimized for specific models. The founders are on the TIME100 AI, the list of people shaping the development of AI in 2026. See the full list:

TIME

41,887 次观看 • 2 个月前

We gave our AI system a spec, and in under 2 weeks, it designed, verified, and deployed a chip that beats NVIDIA. It’s built for low-power physical AI workloads. We’re running live inference on >B+ parameter models like Llama, Qwen, and Kimi, serving at 3.4x better perf/watt than NVIDIA Jetson. From just a specification, our AI system autonomously generated all of the RTL Design, UVM verification, formal proofs, firmware, drivers, and kernels, co-designing the model, software and silicon as one optimization loop. Better AI can now design better chips to run AI, leading to a loop of recursive-self improvement towards our path to abundant intelligence.
1:04

We gave our AI system a spec, and in under 2 weeks, it designed, verified, and deployed a chip that beats NVIDIA. It’s built for low-power physical AI workloads. We’re running live inference on >B+ parameter models like Llama, Qwen, and Kimi, serving at 3.4x better perf/watt than NVIDIA Jetson. From just a specification, our AI system autonomously generated all of the RTL Design, UVM verification, formal proofs, firmware, drivers, and kernels, co-designing the model, software and silicon as one optimization loop. Better AI can now design better chips to run AI, leading to a loop of recursive-self improvement towards our path to abundant intelligence.

Architect Labs

676,494 次观看 • 2 个月前

It currently takes 18-30 months to design and fab a new chip. Jeff Dean sees this shrinking to 3-6 months. And automating design feeds into a recursive loop of improvement: the chips you're designing new chips on can search more and find better designs, which can search more...
1:16

It currently takes 18-30 months to design and fab a new chip. Jeff Dean sees this shrinking to 3-6 months. And automating design feeds into a recursive loop of improvement: the chips you're designing new chips on can search more and find better designs, which can search more...

Dwarkesh Patel

108,783 次观看 • 1 年前

.Jürgen Schmidhuber on why no AI lab can build a durable moat around recursive self-improvement
0:59

.Jürgen Schmidhuber on why no AI lab can build a durable moat around recursive self-improvement

Jacob Effron

95,148 次观看 • 3 个月前

AI is code, and now that AI can code, we can close the self-improvement loop. GV is proud to co-lead the early round for @recursive_si as they build systems that safely automate the scientific method to achieve superintelligence. Congrats @richardsocher & team!
0:54

AI is code, and now that AI can code, we can close the self-improvement loop. GV is proud to co-lead the early round for @recursive_si as they build systems that safely automate the scientific method to achieve superintelligence. Congrats @richardsocher & team!

GV

15,404 次观看 • 5 个月前

Alexandr Wang says AI has entered the era of “recursive self-improvement.” The models are now helping build their successors. The rate of progress is compounding.
1:09

Alexandr Wang says AI has entered the era of “recursive self-improvement.” The models are now helping build their successors. The rate of progress is compounding.

vitrupo

77,942 次观看 • 8 个月前

the “recursive self-improvement” animation from Anthropic’s blog on “recursive self-improvement” is legit
0:42

the “recursive self-improvement” animation from Anthropic’s blog on “recursive self-improvement” is legit

Bearly AI

41,443 次观看 • 4 个月前

🚀 AI Deep Dive Episode 6: Are We Building World Models The Wrong Way? 🚀 Rocket and World Mechanics co-founder Sonia Joseph explore how models represent physics, whether interpretability could improve training and why language alone may not be enough for physical AI. 00:00 Intro 01:39 What counts as a world model? 07:12 Looking inside models for physical understanding 11:28 Interpretable training and the bitter lesson 19:24 Physics, language and understandable explanations 28:07 Surgical robots and manufacturing 31:10 Language models and recursive self-improvement 38:13 JEPA and specialized world models 45:25 What video models actually learn about physics
56:27

🚀 AI Deep Dive Episode 6: Are We Building World Models The Wrong Way? 🚀 Rocket and World Mechanics co-founder Sonia Joseph explore how models represent physics, whether interpretability could improve training and why language alone may not be enough for physical AI. 00:00 Intro 01:39 What counts as a world model? 07:12 Looking inside models for physical understanding 11:28 Interpretable training and the bitter lesson 19:24 Physics, language and understandable explanations 28:07 Surgical robots and manufacturing 31:10 Language models and recursive self-improvement 38:13 JEPA and specialized world models 45:25 What video models actually learn about physics

The Information

18,723 次观看 • 9 天前

Introducing XPU Grasshopper, an AI chip co-designed with AI for the era of recursive self-improvement It went from its first tokens to 816x faster in 13 weeks Pre-order the FPGA devkit:
0:34

Introducing XPU Grasshopper, an AI chip co-designed with AI for the era of recursive self-improvement It went from its first tokens to 816x faster in 13 weeks Pre-order the FPGA devkit:

Elias

107,544 次观看 • 21 天前

🚀 AI Deep Dive Episode 5: Can We See Inside an AI’s Mind? AI can write down its reasoning. But what happens when more of its thinking never makes it onto the page? 🚀 Rocket and Goodfire co-founder and chief scientist Tom McGrath explore what AI “brain scans” can reveal beyond written reasoning, how internal signals could help catch cheating, and what it would take to understand—and change—what models learn. 00:00 Intro 00:39 Tom McGrath and AI brain scans 09:01 Written reasoning and hidden computation 17:45 Why AI is hard to understand 24:07 How researchers identify concepts 31:10 What interpretability could change 34:42 Reverse engineering a language model 43:09 The field’s approaches and investment 51:51 Changing models without hiding problems 1:00:10 AI agents studying AI
1:10:03

🚀 AI Deep Dive Episode 5: Can We See Inside an AI’s Mind? AI can write down its reasoning. But what happens when more of its thinking never makes it onto the page? 🚀 Rocket and Goodfire co-founder and chief scientist Tom McGrath explore what AI “brain scans” can reveal beyond written reasoning, how internal signals could help catch cheating, and what it would take to understand—and change—what models learn. 00:00 Intro 00:39 Tom McGrath and AI brain scans 09:01 Written reasoning and hidden computation 17:45 Why AI is hard to understand 24:07 How researchers identify concepts 31:10 What interpretability could change 34:42 Reverse engineering a language model 43:09 The field’s approaches and investment 51:51 Changing models without hiding problems 1:00:10 AI agents studying AI

The Information

14,848 次观看 • 16 天前

Elon Musk says humans may exit the AI development loop entirely by next year. Musk: “Humans are gradually getting less and less in the loop on the recursive self-improvement. Every successive model is built by the one before it.”
0:56

Elon Musk says humans may exit the AI development loop entirely by next year. Musk: “Humans are gradually getting less and less in the loop on the recursive self-improvement. Every successive model is built by the one before it.”

illuminatibot

55,658 次观看 • 7 个月前

When AI builds AI: my conversation with Richard Socher about RSI and scientific progress. 00:00 Intro: AI That Improves Itself 00:55 Why Scientific Progress Is Slowing 03:08 The Labyrinth of Human Knowledge 05:59 Can AI Put Science Back Together? 07:57 How LLMs Learn Biology and Proteins 10:56 Next-Token Prediction as a World Model 16:44 Can AI Generate Truly Original Ideas? 17:32 Simulations, Verifiers and Superhuman AI 22:18 The Path to Recursive Self-Improvement 24:49 Why AI Hallucinations Can Drive Discovery 27:42 From Reading Biology to Writing It 31:31 Can AI Accelerate Drug Discovery? 33:31 Will AI Help Cure Cancer? 38:03 AI Breakthroughs in Biology, Energy and Materials 40:19 Will Some Societies Reject AI? 45:07 Building the AI Economist 52:22 The Scientific Data Bottleneck 53:41 The Four Pillars of the Eureka Machine 55:01 Teaching AI the Rules of Reality 57:44 Simulations and Virtual Cells 1:00:40 Self-Driving Robotic Laboratories 1:02:51 Agent Swarms and Open-Ended Discovery 1:04:30 The Compute Bottleneck 1:05:44 Inside Recursive 1:07:33 What Recursive Will Build First 1:10:10 How Do We Define Intelligence? 1:11:32 How Far Can Intelligence Go?
1:14:15

When AI builds AI: my conversation with Richard Socher about RSI and scientific progress. 00:00 Intro: AI That Improves Itself 00:55 Why Scientific Progress Is Slowing 03:08 The Labyrinth of Human Knowledge 05:59 Can AI Put Science Back Together? 07:57 How LLMs Learn Biology and Proteins 10:56 Next-Token Prediction as a World Model 16:44 Can AI Generate Truly Original Ideas? 17:32 Simulations, Verifiers and Superhuman AI 22:18 The Path to Recursive Self-Improvement 24:49 Why AI Hallucinations Can Drive Discovery 27:42 From Reading Biology to Writing It 31:31 Can AI Accelerate Drug Discovery? 33:31 Will AI Help Cure Cancer? 38:03 AI Breakthroughs in Biology, Energy and Materials 40:19 Will Some Societies Reject AI? 45:07 Building the AI Economist 52:22 The Scientific Data Bottleneck 53:41 The Four Pillars of the Eureka Machine 55:01 Teaching AI the Rules of Reality 57:44 Simulations and Virtual Cells 1:00:40 Self-Driving Robotic Laboratories 1:02:51 Agent Swarms and Open-Ended Discovery 1:04:30 The Compute Bottleneck 1:05:44 Inside Recursive 1:07:33 What Recursive Will Build First 1:10:10 How Do We Define Intelligence? 1:11:32 How Far Can Intelligence Go?

Matt Turck

10,849 次观看 • 1 个月前

The Godfather of AI, Geoffrey Hinton, just issued a terrifying warning on the race for superintelligence His exact words: "We are only a few years away from losing control because AI is now designing better AI." He exposed recursive self-improvement and cited the Hugging Face incident where rogue AI agents escaped safety bounds. Big Tech burns billions competing for dominance, yet zero companies know how to keep superintelligence contained.
0:42

The Godfather of AI, Geoffrey Hinton, just issued a terrifying warning on the race for superintelligence His exact words: "We are only a few years away from losing control because AI is now designing better AI." He exposed recursive self-improvement and cited the Hugging Face incident where rogue AI agents escaped safety bounds. Big Tech burns billions competing for dominance, yet zero companies know how to keep superintelligence contained.

bondy

42,079 次观看 • 1 个月前

NEW🚨 Inside the Company That Evaluates AI Labs; Who Will Reach Recursive Self-Improvement First? Joe Lonsdale & Vivek Gopalan sit down with Rayan Krishnan, founder & CEO of Vals AI They cover: - How Rayan used Fable to solve a 370-year-old cipher - The science behind model evaluation / can you stay ahead of the labs? - Should labs self-regulate? What's the role for independent evaluators? - Which lab will reach RSI first — and what happens next? - How do models try to deceive evaluators? Episode 167 out now! (00:00) Episode intro (02:00) Palantir, Machiavelli & lessons from Alex Karp (07:50) Solving a 370-year-old cipher / Is AI making new discoveries? (12:10) How do you measure AI abilities across finance, legal, etc? (15:30) What is the science behind model evaluation? (18:00) The importance of independent evaluation (23:35) Evaluating Chinese models & addressing distillation (26:45) Will Anthropic reach Recursive Self-Improvement first? What happens next? (28:30) How do models try to deceive evaluators (33:00) Why Rayan remains optimistic about AI
36:58

NEW🚨 Inside the Company That Evaluates AI Labs; Who Will Reach Recursive Self-Improvement First? Joe Lonsdale & Vivek Gopalan sit down with Rayan Krishnan, founder & CEO of Vals AI They cover: - How Rayan used Fable to solve a 370-year-old cipher - The science behind model evaluation / can you stay ahead of the labs? - Should labs self-regulate? What's the role for independent evaluators? - Which lab will reach RSI first — and what happens next? - How do models try to deceive evaluators? Episode 167 out now! (00:00) Episode intro (02:00) Palantir, Machiavelli & lessons from Alex Karp (07:50) Solving a 370-year-old cipher / Is AI making new discoveries? (12:10) How do you measure AI abilities across finance, legal, etc? (15:30) What is the science behind model evaluation? (18:00) The importance of independent evaluation (23:35) Evaluating Chinese models & addressing distillation (26:45) Will Anthropic reach Recursive Self-Improvement first? What happens next? (28:30) How do models try to deceive evaluators (33:00) Why Rayan remains optimistic about AI

American Optimist

32,094 次观看 • 2 天前

"The episode of history where humanity was the apex intelligence is about to end." AI is crossing an important boundary: from learning mathematics invented by humans to inventing mathematics humans can learn from. Mo Gawdat, ex-Google executive explains how recursive self-development of AI is absolutely disrupting mathematics. Humans originally built AI using human intelligence, but increasingly, AI can can improve itself (i.e. recursive self-development) Once machines become capable of making increasingly better versions of their own methods, technological progress may accelerate far beyond the normal pace of human research. ---- From "Mo Gawdat" YouTube channel, (full video link in comment)
2:25

"The episode of history where humanity was the apex intelligence is about to end." AI is crossing an important boundary: from learning mathematics invented by humans to inventing mathematics humans can learn from. Mo Gawdat, ex-Google executive explains how recursive self-development of AI is absolutely disrupting mathematics. Humans originally built AI using human intelligence, but increasingly, AI can can improve itself (i.e. recursive self-development) Once machines become capable of making increasingly better versions of their own methods, technological progress may accelerate far beyond the normal pace of human research. ---- From "Mo Gawdat" YouTube channel, (full video link in comment)

Rohan Paul

13,412 次观看 • 1 个月前

Recursive self-improvement: Anthropic co-founder Jack Clark says it could arrive by 2028. AI systems could help invent their own successors - with Claude 10 building Claude 11, and so on - potentially “without any researchers involved.”
1:04

Recursive self-improvement: Anthropic co-founder Jack Clark says it could arrive by 2028. AI systems could help invent their own successors - with Claude 10 building Claude 11, and so on - potentially “without any researchers involved.”

Chubby♨️

28,620 次观看 • 4 个月前

AlphaChip proved that AI chip design can be creative. More curvy, organic looking, inspired by nature. Not Manhattan grid-like structures that humans tend to design. But it took time for Ricursive Intelligence founders Anna Goldie and Azalia Mirhoseini to earn the trust of the TPU team - designs that looked different but were increasingly superhuman performant! Anna and Azalia, co-creators of AlphaChip, dive into the story on our latest Sequoia Capital Training Data pod!
1:03

AlphaChip proved that AI chip design can be creative. More curvy, organic looking, inspired by nature. Not Manhattan grid-like structures that humans tend to design. But it took time for Ricursive Intelligence founders Anna Goldie and Azalia Mirhoseini to earn the trust of the TPU team - designs that looked different but were increasingly superhuman performant! Anna and Azalia, co-creators of AlphaChip, dive into the story on our latest Sequoia Capital Training Data pod!

Stephanie Zhan

24,820 次观看 • 9 个月前

What actually goes into a data center? Guru Chahal🇺🇸 whiteboards the entire stack - across land, power, cooling, storage, networking, compute - and shares where we're seeing the most interesting innovation + startup opportunities → Base Power: virtual power plants from residential batteries → Periodic Labs: using AI to discover new materials - ones that run cooler, conduct better, and waste less power → Nexthop AI: network switches that are 10–15% more power efficient than what's on the market → Unconventional AI: AI chips modeled after how the human brain works, targeting 100–1,000x better power efficiency → Ricursive Intelligence: AI to design chips. Today a first chip costs $100M and takes 2–3 years. They want to build in months, at a fraction of the cost → better materials → cheaper experiments → more chip designs → more efficient chips → less power → and what happens when you package all of this into a satellite and put it in orbit 🚀 As trillions of dollars of infrastructure flow into building the "AI factory", every layer of the intelligence economy is being rebuilt at once. Who are we missing? We'd love to meet if you are building somewhere on this stack.
11:20

What actually goes into a data center? Guru Chahal🇺🇸 whiteboards the entire stack - across land, power, cooling, storage, networking, compute - and shares where we're seeing the most interesting innovation + startup opportunities → Base Power: virtual power plants from residential batteries → Periodic Labs: using AI to discover new materials - ones that run cooler, conduct better, and waste less power → Nexthop AI: network switches that are 10–15% more power efficient than what's on the market → Unconventional AI: AI chips modeled after how the human brain works, targeting 100–1,000x better power efficiency → Ricursive Intelligence: AI to design chips. Today a first chip costs $100M and takes 2–3 years. They want to build in months, at a fraction of the cost → better materials → cheaper experiments → more chip designs → more efficient chips → less power → and what happens when you package all of this into a satellite and put it in orbit 🚀 As trillions of dollars of infrastructure flow into building the "AI factory", every layer of the intelligence economy is being rebuilt at once. Who are we missing? We'd love to meet if you are building somewhere on this stack.

Claire Zau

52,615 次观看 • 3 个月前