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.

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

5,096 subscribers

531,806 views • 10 months ago •via X (Twitter)

Education Health & Wellness Science & Technology

Anya Rossi• Live Now

Private livecam show

34 Comments

Jeff Dean10 months ago

Congrats! 🎉

Ricursive Intelligence10 months ago

We are fortunate to partner with investors who share our vision, including @Sequoia (@stephzhan), Modern Capital (@SuhailNimji, @jml_campbell), 49Palms (Madhavan Ramanujam, Joshua Bloom), and @StrikerVP (@maxgazor, @brianzhan1).

Ricursive Intelligence10 months ago

We've built an exceptional team and we are growing: @annadgoldie, @Azaliamirh, @PakJiwoo, @esonghori2, @yclueecs, @DZhang50, @hao9chen. Join us at [email protected].

Steven Xiao10 months ago

incredible team. grateful to have been a part of this launch video

Misha Laskin10 months ago

Congrats! So excited for you

muhtasham10 months ago

congrats @DZhang50 massive W

Holly Peck10 months ago

love this

tuomo10 months ago

i have been planning a post on how nano banana’s version of me as a woman looks like someone who completes her phd, lands a job at deepmind, spins off a company 🤣🤣 don’t know where from this stemmed from but apparently there was a point

AndyXAndersen10 months ago

Nice in theory. In practice, chip design is slow and hard for a reason. Chips are complex, and the pipeline of tools is immense. Even some incremental improvement would be huge and hard to make.

Leonard Tang10 months ago

congrats!!!!!

Antoni10 months ago

It would be fun to someday see a model running the full lithography-deposition cycle on a wafer and getting immediate feedback in close to real-time

Dom10 months ago

Congrats 🫡

Montlake Studios10 months ago

Genuinely a generational team @RicursiveAI. Thank you for trusting us with this project.

Adam Roberts10 months ago

Amazing to see -- congratulations!

✨ nenu10 months ago

It all cooks down to this

Ch🅰️rtradamus 🔋10 months ago

This is sick, best of luck! Looking forward to the greatest advancements related to AI yet to come. Those in fields such as robotics, space, and medicine. The ones that can truly revolutionize our world. Your technology could accelerate that. So many sidetracked by shiny creative AI tools, they disregard the real impact we have yet to ascertain.

ncubelabs10 months ago

Holy smokes 🤯

Today in AI10 months ago

That recursive self-improvement loop is fundamentally driving hardware-software co-design to new levels of integration. The resulting interplay is key to unlocking unprecedented efficiency gains for autonomous systems.

pr337h4m10 months ago

RISCursive was right there 😭😭

Synaptic10 months ago

I'm excited to see what's to come.

Meet10 months ago

gpu designing better gpu!!!

Jacob10 months ago

congrats!

MechatronicsEngineer10 months ago

Excited to see Ricursive push the frontier.

max iuliani ハリ🌐10 months ago

Very cool. Closing the loop between models and the chips they run on is a huge step. From the HARI side we’re obsessed with the complementary problem: a protective “nervous system” that watches those loops for risk, drift and misuse as they scale. #HARI

Nick Sotto10 months ago

Computers designing better computers...

Kazim Haidari10 months ago

Congrats 👏

Akshyae Singh10 months ago

Who built your demo video?

Pablo Pablo10 months ago

Propuesta potente, pero el gran reto será romper los límites prácticos del hardware. ¿Buscáis talento con experiencia real en optimización de chips para ese ciclo recursivo?

Joat Mone (Athreya) 🇮🇳 🇺🇸 🇪🇺10 months ago

Hey you guys misspelled the name of your company

RStorm 202310 months ago

Recursive AI-chip co-design is a major leap. With this caliber of expertise, we could see real movement toward AGI and a new layer of innovation.

imigry10 months ago

@grok How many companies are already doing this and where was this first done?

SOHM10 months ago

very cool!!

Glory.eth| AI automation*,*10 months ago

This is amazing, I love the project. Executing chips designs faster and vertical integration for chips design. Applausible!

Carlton Mackall10 months ago

Welcome to the internet! My big questions (okay actually just what I would do if I could work on chip designs) Do you have a world model for chips yet? also, what's your timescale for producing optical chips?

Related Videos

🚀 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 Is Recursive Self-Improvement? – Can AI Automate Chip Design? – Co-Designing AI and Hardware – Beyond Today’s Models and Chips – How Much More Efficient Could AI Get? – What Happens When the Loop Accelerates? – Can Alignment Keep Up? – Should We Slow AI Progress?

🚀 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 Is Recursive Self-Improvement? – Can AI Automate Chip Design? – Co-Designing AI and Hardware – Beyond Today’s Models and Chips – How Much More Efficient Could AI Get? – What Happens When the Loop Accelerates? – Can Alignment Keep Up? – Should We Slow AI Progress?

The Information

17,374 views • 21 days ago

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.

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 views • 3 months ago

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

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

Jacob Effron

95,148 views • 3 months ago

“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:

“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 views • 1 month ago

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.

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

675,632 views • 1 month ago

Cyber defense is entering the superintelligence era. ⚡ Introducing the CrowdStrike Cyber Superintelligence Lab, the first frontier AI research organization built for cyberdefense and AI safety. The Lab delivers CrowdStrike SafeMind, a family of purpose-built security models and harnesses. SafeMind runs as one system designed to deliver AI safety: an offensive model that finds the attack path, a defensive model that closes it, and the harnesses that operate both in the same loop. Built using NVIDIA Nemotron open models and using CoreWeave’s AI Cloud for training and inference, SafeMind goes beyond AI that simply identifies threats, it defeats them. Learn more:

Cyber defense is entering the superintelligence era. ⚡ Introducing the CrowdStrike Cyber Superintelligence Lab, the first frontier AI research organization built for cyberdefense and AI safety. The Lab delivers CrowdStrike SafeMind, a family of purpose-built security models and harnesses. SafeMind runs as one system designed to deliver AI safety: an offensive model that finds the attack path, a defensive model that closes it, and the harnesses that operate both in the same loop. Built using NVIDIA Nemotron open models and using CoreWeave’s AI Cloud for training and inference, SafeMind goes beyond AI that simply identifies threats, it defeats them. Learn more:

CrowdStrike

58,338 views • 1 month ago

New pod w/ Dwarkesh Patel: AI Continuous Improvement, Intelligence Explosion, Memory, Frontier Lab Competition, & Plenty More Full episode --> Spotify: Apple: YouTube: Etc.

New pod w/ Dwarkesh Patel: AI Continuous Improvement, Intelligence Explosion, Memory, Frontier Lab Competition, & Plenty More Full episode --> Spotify: Apple: YouTube: Etc.

Alex Kantrowitz

18,803 views • 1 year ago

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:

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

107,025 views • 12 days ago

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.”

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 views • 7 months ago

Eric Schmidt says today's AI is human-prompted "special intelligence," not true general intelligence The real danger begins when systems start initiating ideas on their own through recursive self-improvement "the point at which we give agency to computers is a mistake"

Eric Schmidt says today's AI is human-prompted "special intelligence," not true general intelligence The real danger begins when systems start initiating ideas on their own through recursive self-improvement "the point at which we give agency to computers is a mistake"

Haider.

25,751 views • 11 months ago

Introducing Contra Labs. The first frontier data and evaluation lab for Creative AI.

Introducing Contra Labs. The first frontier data and evaluation lab for Creative AI.

ben

451,002 views • 6 months ago

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...

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 views • 1 year ago

Demis Hassabis says the next frontier is AI self-improvement — and coding is where it starts, because the outputs are verifiable and synthetic data is limitless That's a compounding flywheel, and every major lab is betting on it "but raw capability without reliability is a liability"

Demis Hassabis says the next frontier is AI self-improvement — and coding is where it starts, because the outputs are verifiable and synthetic data is limitless That's a compounding flywheel, and every major lab is betting on it "but raw capability without reliability is a liability"

Haider.

15,428 views • 4 months ago

.swyx and I sat down at AIE to talk about AGI, frontier lab strategy, chips, Fable, government, and AI existentialism

.swyx and I sat down at AIE to talk about AGI, frontier lab strategy, chips, Fable, government, and AI existentialism

Matthew Berman

17,214 views • 3 months ago

Excited to release PostTrainBench v1.0! This benchmark evaluates the ability of frontier AI agents to post-train language models in a simplified setting. We believe this is a first step toward tracking progress in recursive self-improvement 🧵:

Excited to release PostTrainBench v1.0! This benchmark evaluates the ability of frontier AI agents to post-train language models in a simplified setting. We believe this is a first step toward tracking progress in recursive self-improvement 🧵:

Karina

207,348 views • 7 months ago

Elon Musk: We're in the hard takeoff right now. Every successive AI model is built by the one before it. Peter Diamandis: Do you see Grok doing recursive self-improvement at this point? And how? What's the timeline for AGI and ASI? Give us a sense of that. Elon Musk: Yeah, I think we've been in recursive improvement for a while here. If you mean by, like, recursive self-improvement without a human in the loop, is that what you mean? Peter Diamandis: I do. Elon Musk: On the AI software side, I mean, humans are gradually getting less and less in the loop on the recursive self-improvement. So every successive model is built by the one before it. So that is happening to a large degree, but it's not yet fully automated. It may be there end of this year, but not later than next year. Peter Diamandis: Do you see a hard takeoff at that point? Elon Musk: We're in the hard takeoff. Peter Diamandis: Okay. Elon Musk: Right now, yes. I mean, at this point I go to sleep, there's some massive AI breakthrough, and when I wake up, there's another one. Yes. Yeah, it's hard to keep track, honestly. So, yeah, it's a bit of a head spinner. Moonshots with Peter Diamandis (Abundance Summit 2026), March 18, 2026.

Elon Musk: We're in the hard takeoff right now. Every successive AI model is built by the one before it. Peter Diamandis: Do you see Grok doing recursive self-improvement at this point? And how? What's the timeline for AGI and ASI? Give us a sense of that. Elon Musk: Yeah, I think we've been in recursive improvement for a while here. If you mean by, like, recursive self-improvement without a human in the loop, is that what you mean? Peter Diamandis: I do. Elon Musk: On the AI software side, I mean, humans are gradually getting less and less in the loop on the recursive self-improvement. So every successive model is built by the one before it. So that is happening to a large degree, but it's not yet fully automated. It may be there end of this year, but not later than next year. Peter Diamandis: Do you see a hard takeoff at that point? Elon Musk: We're in the hard takeoff. Peter Diamandis: Okay. Elon Musk: Right now, yes. I mean, at this point I go to sleep, there's some massive AI breakthrough, and when I wake up, there's another one. Yes. Yeah, it's hard to keep track, honestly. So, yeah, it's a bit of a head spinner. Moonshots with Peter Diamandis (Abundance Summit 2026), March 18, 2026.

ELON CLIPS

19,692 views • 3 months ago

A challenge from Daniel Kokotajlo to AI company employees who believe recursive self-improvement will kick off in the next couple of years:

A challenge from Daniel Kokotajlo to AI company employees who believe recursive self-improvement will kick off in the next couple of years:

Palisade Research

33,462 views • 1 month ago

🚀 Introducing Community Benchmarks on Kaggle! As AI evolves at an unprecedented pace, measuring intelligence requires more than a few AI research labs alone – it requires the imagination and collective expertise of the global community. That’s why we’re launching Community Benchmarks. It lets you build, run, and share custom AI benchmarks that are evaluated on the leading AI models with transparent, reproducible results. Learn more:

🚀 Introducing Community Benchmarks on Kaggle! As AI evolves at an unprecedented pace, measuring intelligence requires more than a few AI research labs alone – it requires the imagination and collective expertise of the global community. That’s why we’re launching Community Benchmarks. It lets you build, run, and share custom AI benchmarks that are evaluated on the leading AI models with transparent, reproducible results. Learn more:

Kaggle

54,695 views • 9 months ago

Introducing ScienceBuddy — a free workspace for scientific agents that improve through researcher collaboration. Use GPT-6 in ScienceBuddy at no cost. GPU-accelerated, and fused with the JEV framework. 🧵 Two loops: 🔹 Inner loop — refines the agent harness 🔹 Outer loop — trains the model with rubric-guided RL Together: Recursive-in-Recursive Self-Improvement. ScienceBuddy explores how scientific agents can improve through sustained collaboration with researchers. 🔬Try it free: #ScienceBuddy #PhAILabs #AI4Science

Introducing ScienceBuddy — a free workspace for scientific agents that improve through researcher collaboration. Use GPT-6 in ScienceBuddy at no cost. GPU-accelerated, and fused with the JEV framework. 🧵 Two loops: 🔹 Inner loop — refines the agent harness 🔹 Outer loop — trains the model with rubric-guided RL Together: Recursive-in-Recursive Self-Improvement. ScienceBuddy explores how scientific agents can improve through sustained collaboration with researchers. 🔬Try it free: #ScienceBuddy #PhAILabs #AI4Science

PhAI Labs

1,254,161 views • 26 days ago

Introducing Project Ceiba, a groundbreaking collaboration between #AWS and NVIDIA, that aims to push the boundaries of #AI by constructing the largest AI supercomputer in the cloud. ☁️ ⚙️ 💻 Learn more. ➡️

Introducing Project Ceiba, a groundbreaking collaboration between #AWS and NVIDIA, that aims to push the boundaries of #AI by constructing the largest AI supercomputer in the cloud. ☁️ ⚙️ 💻 Learn more. ➡️

Amazon Web Services

16,205 views • 2 years ago