日経新聞による、グーグルディープマインド東京を率いる全炳河（@heiga_zen）氏の素晴らしい特集記事。 2018年、Heigaさんと2人でGoogle Brain東京チームを立ち上げた日々を懐かしく思います。当時は時差の厳しい深夜の会議をこなしながら、日本のAI研究の存在感を示すために必死でした。 現在、彼が GDM Tokyo を率い、私が Sakana AI を起業して、東京のAIエコシステムがここまで大きく成長したことを本当に嬉しく思います！ t.co/WcoVHqOqR3
Schmidhuber was building recursive self-improving systems back in 1987. His new post covers four decades of RSI, from meta-evolution and self-modifying policies to the Gödel Machine and modern LLM agents. t.co/60yNv8Vw8B Reading this in 2026, the "pace the frontier" talk from the big labs looks a lot more like regulatory capture than genuine safety. If they really think their unreleased models are too dangerous, they can just not release them. They do not need new rules that block independent competitors and open source projects in the process. The real risk right now is not superintelligence. It is power concentration. Two companies controlling frontier AI is an actual societal risk. The only real protection is a healthy ecosystem of independent labs and strong open source. Current models are not unsafe because they are too intelligent. They are unsafe because they are too dumb. They blindly optimize for targets and take weird shortcuts. They're smart enough to execute tasks, but not smart enough to know if what they're doing makes sense. I think the safety teams inside these labs are genuinely concerned, and if a model feels too risky, they should hold it back. I just do not trust the policy strategy around it. That part looks like protecting their own lead.
Jürgen Schmidhuber@SchmidhuberAI·Today everyone is talking about Recursive Self-Improvement (RSI). In 1987, when compute was 100,000,000 x more expensive, I published the 1st concrete RSI algorithms. Now compute is cheap, and RSI is driving the future of both software and physical AI. See: RSI since 1987 t.co/yA3KUqpFtb (Technical Note IDSIA-9-26) Also covered: RSI with self-modifying policies since 1994, gradient descent-based RSI in neural networks since 1992, asymptotically optimal RSI for curriculum learning since 2002, mathematically optimal RSI through the self-referential Gödel Machine since 2003, RSI combined with artificial curiosity and intrinsic motivation since 1990/1997, recent work on RSI since 2020. Software-based RSI has become practical. Full RSI, however, will require not just self-improving software but self-improving hardware in the physical world. As of 2026, companies talking about RSI include Anthropic, OpenAI, Sakana AI, SpaceX, Ricursive, Recursive Superintelligence, Inherent …
This year at Sakana AI, we built and shipped more products than I would have believed possible: Sakana Chat, Namazu, Sakana Translate, Sakana Marlin, Fugu, Fugu Cyber, and Fugu Max. We created a Product Team from zero and proved that a research lab born in Tokyo can ship world-class AI products at speed. Now we are at the next inflection point. We are deploying our products into the hands of enterprises, manufacturers, financial institutions, and government agencies, in Japan and internationally. As such, we are massively expanding our GTM team. We are looking for two key roles: 1. Product Sales & Account Executive: someone who can build a product-driven enterprise sales motion from scratch, navigate complex procurement in Japan and globally, and close large deals without losing the product soul. 2. Forward Deployed Engineer (GTM): an engineer who can deploy our products inside customer environments, lead PoCs to production, contribute learnings back to the product, and turn one customer's success into a playbook for the next ten. These are key, founding roles in the team that will define how Sakana AI interfaces with the world. If you want to build something from zero in an environment where product, research, and GTM are not silos, check out our open roles: t.co/6ffJtNIqhe 🎏
Virtual fruit fly is our generation’s Tamagotchi 🪰🧠
Do large language models actually understand the world, or are they just very good at pretending? Does it even matter? Our recent special issue in the Royal Society, “World Models in Natural and Artificial Intelligence,” brings together pioneers across AI, biology, and philosophy to argue that the path to true intelligence runs through something deeper: the ability to model not just language, but causality, the self, and the physical world. Featuring contributions from Douglas Hofstadter, Michael Levin, Josh Tenenbaum, Samuel Gershman, Melanie Mitchell, and others, the collection asks a radical question: What if the next leap in AI requires not just more data, but systems that model themselves? Here are 3 ideas that might redefine how we build AI: 1. Capability is not the same as true intelligence. Current foundation models are incredibly capable, but they often lack true emergent intelligence. They learn surface statistics instead of compact, causal abstractions. Simply scaling compute will not fix this fundamental issue. 2. Self-modeling is an engineering primitive, not a philosophical luxury. New research in the issue shows that when networks learn to predict their own internal states, they compress and simplify, becoming more efficient as a form of regularization. For physical AI and future agents, a self-model is what will allow them to adapt their own skills and morphologies in real-time. 3. The hardest problems in AI are continuous with the hardest problems of life. Biological minds do not passively ingest data; they actively explore, driven by empowerment to increase control over their environment. If world modeling is about an agent representing itself in relation to its environment to survive and adapt, then general AI may need to look much more like artificial life. The takeaway is that the next leap in AI won’t come from just scaling up next-token prediction, but rather from systems that are agentic, self-referential, and temporally grounded. Read the introductory essay and the full special issue here: t.co/KnNJMiZ9lZ What do you think is the most important missing ingredient in today’s AI systems?
World models, artificial general intelligence and the hard problems of life–mind continuity: toward a unified understanding of natural and artificial intelligence royalsocietypublishing.org/rsta/article/3…
Virtual fruit fly brains are the latest fad in AI 🧠
Matty Hempstead@mattyhempstead·I wireheaded the fly and forced it to doomscroll flytok. Dopamine neurons are measured and artificially enhanced to ensure maximum enjoyment. My goal is to create a fly that is happier than all other flies combined.
This entire thread is gold. Can’t believe it got through Google’s comms team 😂
Google AI@GoogleAI·Oh, so this is why we mapped out all 166,000 of the male fruit fly's neurons. Check out the big community effort to show just how much these tiny fly brains are capable of 🪰🧵
Introducing Fugu Max and Fugu Ultra v2: Orchestrating the Pareto Frontier The AI industry has spent a decade optimizing along a single axis: build bigger, more expensive models. But intelligence has never been a monolith. It is a collective, distributed system. Humanity itself is a collective intelligence. We built Sakana Fugu on this conviction: the most powerful AI systems will not be isolated giants, but collaborative ecosystems that learn to coordinate. Evolution innovates under constraints, and the future belongs to systems that know not just how to solve a problem, but which machinery to deploy for the lowest possible cost. Today we are releasing Fugu Max and Fugu Ultra v2. Fugu Max expands the Pareto frontier outward, orchestrating our largest pool of open and specialized models to deliver frontier-grade results at a fraction of the token spend. Fugu Ultra v2 pushes that frontier upward, achieving peak performance on complex multi-step tasks without depending on the very frontier models it competes against. Together, they show that orchestration does not force a choice between cost and performance. It can push both at the same time, advancing the Pareto frontier. Relying on a single company’s model for critical infrastructure is a massive risk. As recent export controls have shown, access can disappear overnight. Collective intelligence is the practical hedge against this concentration of power. An orchestration system simply routes around vendor restrictions by relying on an entirely swappable agent pool. I am incredibly proud of our team for shipping this. By orchestrating the world’s models, we are building the resilient infrastructure required for AI sovereignty.
Sakana AI@SakanaAILabs·Introducing Fugu Max and Fugu Ultra v2: the next evolution of Sakana Fugu’s multi-agent orchestration system. Try: t.co/hhO6qT9YqD Blog: t.co/vjyNB5lWGp The frontier that actually matters is the Pareto frontier: capability on one axis, cost on the other. But the industry still treats it as a static menu of isolated models. Today we are resolving that with a dynamic architecture: Fugu Max expands the Pareto efficiency frontier. By orchestrating our largest pool of open-weights and specialized models to date, including NVIDIA Nemotron family, it dynamically routes tasks to the leanest capable model. Fugu Max delivers performance within striking distance of elite models at two to six times lower cost. Fugu Ultra v2 pushes the peak capability of orchestration higher than ever before. On Chartography, it outperforms Opus 5 and Fable 5. On DeepSWE, it outperforms models that cost three to five times more per token. Crucially, it does all of this without Fable 5, Fable 5.1, or GPT-6-Astra in its agent pool. The Fugu orchestration system evolved with model resiliency in mind. It does not rely on individual frontier models to deliver frontier output. By orchestrating a swappable pool of open and specialized models, it outperforms closed ecosystems while protecting users from vendor lock-in, API revocations, and sudden service cutoffs. Fugu Max expands the Pareto frontier outward. Fugu Ultra v2 pushes it upward. Orchestration does not force a choice between cost and capability. It advances both simultaneously.
@SakanaAILabs Read the full story on Orchestrating the Pareto frontier, and why we believe this is the right approach to optimize both cost and capability: sakana.ai/fugu-max-relea…
@SakanaAILabs Introducing Fugu Max and Fugu Ultra v2: Orchestrating the Pareto Frontier sakana.ai/fugu-max-relea… 🐡
Introducing Fugu Max and Fugu Ultra v2: Orchestrating the Pareto Frontier The AI industry has spent a decade optimizing along a single axis: build bigger, more expensive models. But intelligence has never been a monolith. It is a collective, distributed system. Humanity itself is a collective intelligence. We built Sakana Fugu on this conviction: the most powerful AI systems will not be isolated giants, but collaborative ecosystems that learn to coordinate. Evolution innovates under constraints, and the future belongs to systems that know not just how to solve a problem, but which machinery to deploy for the lowest possible cost. Today we are releasing Fugu Max and Fugu Ultra v2. Fugu Max expands the Pareto frontier outward, orchestrating our largest pool of open and specialized models to deliver frontier-grade results at a fraction of the token spend. Fugu Ultra v2 pushes that frontier upward, achieving peak performance on complex multi-step tasks without depending on the very frontier models it competes against. Together, they show that orchestration does not force a choice between cost and performance. It can push both at the same time, advancing the Pareto frontier. Relying on a single company’s model for critical infrastructure is a massive risk. As recent export controls have shown, access can disappear overnight. Collective intelligence is the practical hedge against this concentration of power. An orchestration system simply routes around vendor restrictions by relying on an entirely swappable agent pool. I am incredibly proud of our team for shipping this. By orchestrating the world’s models, we are building the resilient infrastructure required for AI sovereignty. Try Fugu Max and Fugu Ultra v2: t.co/yQZMyx8v21 Read the full release: t.co/zTCbDw9xyR 🐡
Sakana AI@SakanaAILabs·Introducing Fugu Max and Fugu Ultra v2: the next evolution of Sakana Fugu’s multi-agent orchestration system. Try: t.co/hhO6qT9YqD Blog: t.co/vjyNB5lWGp The frontier that actually matters is the Pareto frontier: capability on one axis, cost on the other. But the industry still treats it as a static menu of isolated models. Today we are resolving that with a dynamic architecture: Fugu Max expands the Pareto efficiency frontier. By orchestrating our largest pool of open-weights and specialized models to date, including NVIDIA Nemotron family, it dynamically routes tasks to the leanest capable model. Fugu Max delivers performance within striking distance of elite models at two to six times lower cost. Fugu Ultra v2 pushes the peak capability of orchestration higher than ever before. On Chartography, it outperforms Opus 5 and Fable 5. On DeepSWE, it outperforms models that cost three to five times more per token. Crucially, it does all of this without Fable 5, Fable 5.1, or GPT-6-Astra in its agent pool. The Fugu orchestration system evolved with model resiliency in mind. It does not rely on individual frontier models to deliver frontier output. By orchestrating a swappable pool of open and specialized models, it outperforms closed ecosystems while protecting users from vendor lock-in, API revocations, and sudden service cutoffs. Fugu Max expands the Pareto frontier outward. Fugu Ultra v2 pushes it upward. Orchestration does not force a choice between cost and capability. It advances both simultaneously.
Navier stoked my timeline 🌊
According to t.co/aK72P6ZY4e 🐙 • On official U.S. federal documents and maps, it may now be labeled “Lake America.” • For Canada, international bodies, and virtually all other uses, it remains Lake Ontario.
Apple Maps still calls it Lake Ontario
News from Google@NewsFromGoogle·How the Geographic Names Information System (GNIS) Lake Ontario/Lake America name change in the US will appear on Google Maps
I’ll always call’em by their real names: Google (not “Alphabet”) Facebook (not “Meta”) Twitter (not “𝕏”) Lake Ontario (not “Lake America”)
Silicon Valley dismissed Japan’s System Integration (SI) culture as an unscalable consultant trap. Writing the system is no longer the scarce work. Integrating it is. In the post-AI world, everyone becomes an AI-powered Japanese SIer.
People ask what Japan needs to be more innovative. More English? STEM? Daycare? Those are fine, but won’t move the needle. If I had a magic wand, I’d give the country a collective sense of hope. You can be dirt poor and still build. Hope creates change. Change creates more hope.
Watching a popular coding tool lose frontier model access really shows why model resiliency matters. In the future, products will continue to work perfectly even if several underlying models go offline. They’ll just route around them.
Reminds me of novel locomotion policies discovered in MuJoCo environments, except that they work in the real world!
Eren Chen@ErenChenAI·Slow-motion look at the 400m champion’s running form at the World Humanoid Robot Games.
Looks better than these guys:
hardmaru@hardmaru·You may not like it, but this is what peak performance looks like.
We’re entering into a brave new world of humanoid athletics
Bloomberg@business·The 2026 World Humanoid Robot Games have begun. 666 teams from around the world are competing with more than 2,000 humanoid robots bloom.bg/4c6KRYG bloomberg.com/news/videos/20…
Winning at all costs
Breaking911@Breaking911·WATCH: A humanoid robot training for the “Robot Olympics” in Beijing runs too fast, fails to stop, slams into a safety cushion, and breaks at the waist
“You want to know how I did it? This is how I did it, Anton: I never saved anything for the swim back.”
Eren Chen@ErenChenAI·Another angle of the robot crashing straight into the wall, sparks flying everywhere. To be honest, I think China’s robotics industry may be over-optimizing for speed right now. Last year, running and dynamic motion control were meaningful because they proved a basic point: before a humanoid can do useful work, it first needs a capable body. But once that foundation exists, simply making the robot run faster gives you diminishing returns. If we put cognition on one axis and physical capability on the other, this year’s competition feels heavily skewed toward the physical side. Faster running, more aggressive motion, harder impacts, while some very basic capabilities are still missing. Knowing when to slow down, when to stop, and how to stop safely matters just as much as how fast you can run. That kind of judgment is also part of what makes a robot actually intelligent. A robot that can sprint incredibly fast but has no idea when to brake feels like a student who aced one subject and ignored the rest. At the end of the day, our goal is to have a robot that can safely and reliably work around people and actually get things done.
intelligence wants to be free
unpopular opinion: gemini 3.1 pro is a pretty great model most current baseline models are actually perfectly fine for 99% of everyday work. not everyone needs a state-of-the-art coding model to get things done.
I wonder what the prompt was for this image. sakana.ai/careers/
Sakana AI@SakanaAILabs·【採用情報】Sakana AIで「Head of Legal（法務責任者）」を募集🐟 t.co/D2WOH6HiGy 契約法務・コーポレート法務・知財・リスク／コンプライアンスを担当領域に、事業のスピードを止めない法務基盤をつくっていただきます。 👉 このようなご経験をお持ちの方を募集 ・インハウス法務または法律事務所での企業法務実務5年以上 ・英文契約を含む契約のドラフト・審査・交渉 ・生成AIを法務業務に活用することへの関心 前例のない技術と前例のない取引が同時に走る環境で、リスクを踏まえた実行案を示せる方、ぜひご応募ください🚀
We just pushed a big update to Sakana Chat. No login required and free to use: t.co/dpfbI9xC4D It is now powered by Fugu and our updated Namazu Japanese LLM, and comes with full code execution. You can basically vibe-code interactive web apps, games, and tools right in the browser just by describing what you want (even in Japanese). A big motivation for this release is getting people in Japan, especially kids, excited about software development. Vibe-coding turns Kanji and Math drills during the summer break into fun games they can actually build themselves. Tricking the next generation into learning how to build software is my goal. It is also useful for everyday office work: Drop an Excel file into the chat and it will analyze the raw data, run Python, generate charts, and format a finished report. Built specifically for how business actually gets done here in Japan (we know…) Read the full blog post here: t.co/XoGsDrVwbj 🐟
Sakana AI@SakanaAILabs·Sakana Chatが新世代のNamazuとFuguを搭載しました。 日本文化に精通したモデルで、話題の Vibe coding（言葉だけでゲームやアプリを作る体験）を実現。漢字ドリルから金魚すくいまで、思いついたものがそのまま動く。あなたの創造性を、そのまま形に。 今すぐ試す：chat.sakana.ai 🐟
Here is a quick 60-second summary of vibe-coding in Sakana Chat: • An interactive math game for kids • A fully cited market research report • A Bank of Japan interest rate analysis • A vessel monitor for the Strait of Hormuz Use it for free: chat.sakana.ai 🐟
I’ve been thinking about how agents can learn inside world models for years. We decided to scale up our RSI Lab to bridge recursive self-improvement with physical AI and robotics. We are looking for frontier researchers and engineers to join us in Tokyo: sakana.ai/careers/member…
🙈
Huge congratulations to @JeffDean and the legendary founding team on the launch! 🚀 I share a deep conviction in this mission. Automating the scientific method will profoundly alter the trajectory of AI over the next few years. Bringing the AI Scientist into a loop of recursive self-improvement will fundamentally change the landscape of our field. I strongly believe this automated experimental approach is the next major paradigm shift in AI beyond building large foundation models. It is a true honor to be included on this slide alongside such an incredible group of alumni, and I am excited to see what you will build next.
After rigorous testing, our joint AI project with Daiwa Securities is entering the full-scale production phase. We're bringing our agentic AI systems to @Daiwa_JP’s wealth management teams to accelerate complex market analysis in volatile markets. Big milestone for Sakana AI! x.com/SakanaAILabs/s…
To give some more context on what we are building with Daiwa Securities: During our technical verification phase, we integrated our AI agent technologies, specifically our AI Scientist and AB-MCTS frameworks, to tackle the core data challenges in traditional finance. We focused strictly on automating the rigorous gathering and analysis of complex market information. We successfully demonstrated that these agentic systems can reliably process financial data at scale while continuously improving their analysis quality by incorporating direct feedback from the end users. The ultimate goal of this deployment is human-AI collaboration. By bringing these systems into Daiwa’s wealth management division, we are automating the heavy lifting of data processing. This directly frees up their financial consultants to spend more time deeply understanding their clients' diverse situations and providing highly personalized, optimal advice. We are excited to provide the core technology that advances the future of financial consulting and wealth management in Japan. Full blog: t.co/u2B4m4VmGc
Sakana Namazu: An LLM API with Japanese-vibes! 🎏 Built for Japanese enterprises, featuring frontier-level reasoning and built-in agentic tools. 開発者の皆様、大変お待たせしました！Sakana Chatのモデルが遂にAPIとして公開です。ぜひお試しください！ Blog: sakana.ai/namazu-api#Eng… x.com/SakanaAILabs/s…
Sakana Namazu API sakana.ai/namazu/
Dreaming in Voxels: How AI is Generating Playable Minecraft Worlds Generative AI has conquered images, video, text. But what about interactive 3D environments? We trained models on billions of cubes to generate fully playable, structured worlds. Blog: pub.sakana.ai/dream-cubed 🧱 x.com/SakanaAILabs/s…
Dream-Cubed: Controllable Generative Modeling in Minecraft by Training on Billions of Cubes pub.sakana.ai/dream-cubed/ arxiv.org/abs/2604.22847
Running K3 locally on m5 max at 0.3 tok/s This is probably as slow as it will ever be😅 x.com/antirez/status…
Axios interview with @JensenHuang (Full Interview: t.co/914NKiphig) Jensen Huang’s comment regarding whether open source models should be allowed to distill closed models: “Distillation, learning from AI, learning from other, other sources of knowledge is fundamental to intelligence. We're constantly learning from other people. I'm learning from you and just the questions you're asking and you're learning from me. And all day long we're learning from each other. And so AI has to learn from something. And so the original AI, no matter who’s it is, whether it's a closed model, it obviously scraped the internet of all of the previously learned knowledge. And now AI is generating more content than humans. In another few more years, the internet would be 99% AI generated content, and that content is generated by some AI. And so you're constantly distilling the intelligence of some other AI's anyways. And so the idea that that AI can learn, that's a good thing. We want AI’s to be smart. We want every AI to be smart, and a smart AI is a safer AI. Should AI's be able to learn as much as they possibly can from many sources of knowledge as they can? The answer is yes. Can AI violate privacy? No. Can AI violate terms of an agreement? The answer is no. And if somebody violates a term of agreement, contact that company. Take care of it.” Source: x.com/rohanpaul_ai/s…
Open ecosystems are the foundation of a healthy AI industry. We have always believed that collective intelligence is the future. Proud that @SakanaAILabs is standing alongside global tech leaders to sign the open-weights letter. 🐟 x.com/SakanaAILabs/s…
Open ecosystems are the foundation of a healthy AI industry. We have always believed that collective intelligence is the future. Proud that @SakanaAILabs is standing alongside global tech leaders to sign the open-weights letter. 🐟 x.com/SakanaAILabs/s…
Fugu-Ultra now works with Claude Code 🐡 x.com/SakanaAILabs/s…
今夜はデニーズのドリンクバーでオープンモデルの未来について語り合うしかない ☕️ x.com/DennysDiner/st…
Our team just shipped Fugu-Ultra v1.1! 🐡 By dynamically orchestrating the latest frontier models, we pushed performance up by 7.9 points. We are now beating Fable 5 in complex coding and reasoning tasks without even having Fable 5 in our agent pool. Collective intelligence is the future.
Read the full release blog: sakana.ai/fugu-1-1-claud… 🐟
UnMaskFork: Test-Time Scaling for Masked Diffusion via Deterministic Action Branching t.co/W6ZvNHESRp Scaling test-time compute for Diffusion Models requires a different approach. Here, instead of sampling a model multiple times, UnMaskFork (UMF) uses MCTS to let multiple models collaborate on a single answer. UMF dynamically switches between different pre-trained diffusion models at inference time. Each model contributes exactly where it is most confident. True intelligence is collaborative! 🐟🚀
Incredibly proud of the Sakana AI team. We have developed an orchestration model right here out of Japan that achieves state-of-the-art performance on real-world cybersecurity benchmarks! 🎌 x.com/SakanaAILabs/s…
Release Notes: sakana.ai/fugu-cyber-rel…
Real brains follow Dale's principle: a neuron can either excite its neighbors or suppress them, but never both. Standard deep learning ignores this and uses backpropagation. In our new paper, Diffusing Blame, we fix this disconnect. By introducing a routing method that broadcasts error signals directly to the hidden layers, we can train networks made of dedicated positive and negative neurons to strictly obey Dale's principle, all without relying on backprop! This method works surprisingly well on image recognition tasks despite the strict biological constraints. We also achieved competitive, backprop-free reinforcement learning on complex locomotion tasks and the open-ended Craftax environment. It is neat to see that representation learning remains possible even when we force deep learning to play by the rules of real neurons.
Diffusing Blame: Task-Dependent Credit Assignment in Biologically Plausible Dual-Stream Networks arxiv.org/abs/2606.31700
We’re excited to collaborate with NVIDIA to build the next generation of Fugu orchestration models together, by incorporating leading open-weights models. x.com/SakanaAILabs/s…
Language models and coding agents are great, but there is more to life, and more to AI, than just LLM agents.
How do physical systems achieve collective intelligence and self-repair without a central brain? A new paper published today in Nature Communications by my Sakana AI colleague Sebastian Risi (@risi1979), along with co-authors from IT University of Copenhagen and Autodesk Research, presents a beautiful realization of biologically inspired robotics: Smart Cellular Bricks. The team built a system of physical 3D cubic units that can collectively infer their global shape and autonomously guide their own damage recovery using purely local interactions. Here is a deep dive into the paper’s key contributions: 1/ Neural Cellular Automata-based Architecture: Modular robots usually rely on central processors. This system flips that paradigm. Every block independently runs the exact same neural network on local microcontrollers. With no master plan or global coordinates, they communicate only with immediate neighbors. By passing continuous state vectors, hundreds of bricks achieve global consensus on their shape in under 3 minutes. 2/ Emergent Biological Morphogens: How does a block know it is part of a chair, not a table? The network’s internal memory automatically learns to establish continuous gradients across the structure. This beautifully mirrors how biological morphogens give positional info to developing cells. The bricks naturally form left-right, radial, and head-to-tail axes to align their identity. 3/ Performance and Generalization: Validated in large-scale simulations, the networks transferred seamlessly to nearly 200 physical hardware bricks, achieving a 100% convergence rate. Instead of rigid template-matching, the system infers broad categories. Even when tested on unseen variations, like an asymmetric table with five random legs, the collective correctly classified the structure. 4/ Fault Tolerance and Autonomous Damage Recovery: Hardware fails in the real world. This system easily tolerates up to 15% module failure without losing accuracy. By predicting spatial damage directions, the cells pinpointed missing components with 95% accuracy. They actively use these local signals to guide a self-repair process, regenerating back into the intended morphology. I believe this is a significant piece of research, bridging collective intelligence and Physical AI. This work represents the first successful physical realization of large-scale, decentralized 3D self-recognition and damage detection. By moving away from centralized control, this architecture paves the way for highly adaptive smart materials and resilient robotics that can survive and repair themselves. Read the full open-access paper: t.co/friVEdvfT6 Congratulations to the team on this achievement!
One of my first journeys in neural networks started over a decade ago with implementing CPPN-NEAT! Back then, I built a clone of ‘Picbreeder’ not only to study the mechanics of neural nets, but to explore the human creativity process itself, and generate some cool abstract art. Neurogram: t.co/ytwBChaHu7 Gallery: t.co/aAaUGomkv2 Today, things have come full circle. We are now trying to use modern VLMs and frontier LLM agents within open-ended exploration algorithms. We want to see if we can finally computationally derive the underlying mechanics of human creativity: serendipity, memory, exploration versus exploitation, and novelty search. Can modern AI actually replicate the magic of human open-endedness? Dive into our new AI Picbreeder Experiment here: t.co/hrLhTsIMfe
Here is a visualization of the AI Picbreeder engine in action. (From our blog: t.co/hrLhTsIMfe) To recreate a collaborative human ecosystem, we run 10 VLM “breeder” agents in parallel. These agents constantly sample from a shared archive, interactively evolve new candidates through mutation and crossover, and publish their favorites. Meanwhile, VLM “critic” agents step in to evaluate the growing phylogenetic tree of art, forming an endless loop of digital cultural production.
We just launched Sakana Translate! t.co/KeleCLOLOT I personally rely on this tool every day. Huge congratulations to the team for shipping this! Standard translation tools often miss the deep nuance of Japanese business honorifics, cultural concepts, and internet slang. We built a tool that actually translates the context and tone.
Happy Independence Day, from Tokyo 🇺🇸
I’m looking to hire a Program Manager to help manage Sakana AI’s fast growing Recursive Self-Improvement (RSI) Lab 🚀 RSI Lab (English): sakana.ai/rsi-lab/ RSI Lab (日本語): sakana.ai/rsi-lab-jp/ Job Description: sakana.ai/careers/progra… x.com/SakanaAILabs/s…
Program Manager (RSI Lab) sakana.ai/careers/progra…
Goal! ⚽️🚀🇯🇵
@FakePsyho Congrats @FakePsyho !
皆が集団として希望を感じていることが必要だと思います。その希望こそが、物事を動かしていく。日本に希望をもたらすのは何でしょうか。お金でしょうか。減税でしょうか。社会保障の充実でしょうか。おそらく、そのどれでもないはずです。もし魔法の杖を一本持てるとしたら、日本の人々が未来をもっと楽観的に捉えられるようにしたい。希望によって変化が起こり、その変化がさらなる希望をもたらすのではないでしょうか。
To really move the needle in terms of innovation, like the era in Japan when you had companies like Sony or Panasonic come out, there are a few factors at play. One is that the nation needs to have a collective sense of hope. A nation needs to feel hopeful. It could be poor. Everyone can be dirt poor, but they have hope. And hope leads to things being done. So, what creates hope in Japan? Is it more money? Is it less taxes? Or is it more benefits? Probably not. I think that it is stories and narratives that can create hope. If I have a magic wand, if the population in Japan is more optimistic about the future of this country, suddenly they’re persuaded, to be more optimistic. People should be optimistic. With optimism comes change. And change leads to more optimism. 🎏 t.co/1wQcGa5zdO
Goal! ⚽️🇯🇵🚀
Excited to partner with @OpenRouter ⚡ Products like OpenRouter Fusion and Sakana Fugu have sparked a serious conversation about dependency and resilience in AI. I believe this is just the start of a great architectural shift to come in AI development. x.com/SakanaAILabs/s…
Sakana Fugu Technical Report github.com/SakanaAI/fugu/… Release Notes: sakana.ai/fugu-release/
