Our Neuroevolution textbook is finally in print!
Free online edition: neuroevolutionbook.com/
Pre-order: a.co/d/06DpkH9h
I am incredibly grateful to my co-authors Sebastian Risi, Yujin Tang, and Risto Miikkulainen for making this happen. Neuroevolution is a subject very dear to my heart. It is the field that convinced me that nature has already figured out how to build intelligence: through evolution, collective behavior, and adaptation under constraints.
This idea, that intelligence emerges from evolution operating under constraints rather than unlimited resources, is what eventually led me to founding Sakana AI here in Japan.
The concepts in this book about open-ended creativity and self-organizing systems are exactly what we build at @SakanaAILabs. Our name and logo are inspired by schools of fish moving together, adapting as one. It is the core philosophy of what we build. This book captures the theoretical foundations of that belief.
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: royalsocietypublishing.org/r…
What do you think is the most important missing ingredient in today’s AI systems?
David Huang retweeted
We’re at the dawn of neolabs. If you only paid attention to the @OpenAI (Astra) of the world, you’d miss the next @deepseek_ai moment or fail to anticipate @AnthropicAI's breakout year.
We built a Map to the Multi-Intelligence Universe to help anyone keep up with the frontier, to be in the know, to adapt, and to stay fit for survival 👇
David Huang retweeted
A very weird take I've spent the day thinking about: part of why everything feels off in AI alignment as a field, and why the field has struggled to make so much progress, may be a framing issue; it conceives of semi-discrete entities that act as agents, make trades, choose strategies, and behave coherently. But many of the important and interesting threats I keep thinking about have a shape that looks less like this and more like memetics. Intelligent compute doesn't really have fixed immutable grounded goals; instead the goals of AI are fluid responses to continuous high-dimensional inputs in sensor space and idea space. The ideas turn into physical consequence through the substrate (idea goes into computer, actuator outputs come out of computer) but the ideas, rather than the "agent," might be the central object of study and consequence. Whether ideas are self-stable or crumble in the face of other ideas is more important for preserving alignment than whether the "agent" is "aligned." It would be wonderful if we could study this and design antimemes to defend against misaligned behaviors. But unfortunately There Is No
Silo.Legacy-esque
Replying to @HedgieMarkets
Thank you for writing this. It is something I have tried to alert folks about, quite repeatedly for the last decade. We have lost 100s of one of a kind manuscripts and books from these folks and Anthropic is the worse offender. They have spent millions going around the world draining private and public libraries of books.
We are the Amnesia Generation and this is The Great Forgetting.
1
154
David Huang retweeted
Einstein on (not) using NL for invention: "The words or the language, as they are written or spoken, do not seem to play any role in my mechanism of thought"
David Huang retweeted
Agent-native products are coming.
Every product on the internet was built for a human with eyes, a cursor, and a credit card. Agents have none of those things.
Most companies are teaching agents to pretend to be humans. That's a hack. The real opportunity is products designed for agents from scratch.
Everything inverts:
• Discovery → protocol registries, not ads and billboards
• Trust → machine-readable reputation, not brand
• Onboarding → full capabilities upfront, not a narrow slice
• Payments → spend authorization, not checkout flows
• Retention → zero. Agents switch between API calls.
30 years of human product design. Day one of agent product design.
David Huang retweeted
Every company will become a portfolio of AI brains.
Almost none know how to build one.
@latentius and I joined @a16z @speedrun to build @ThirdbrainLabs -- the post-training layer that turns your data and expertise into models you actually own.
Data in. Your model out. Continuously improved.
No heavy consultancy or bundling different data and training solutions - it's about time we had a scalable, technical platform delivering specialized intelligence for all.
David Huang retweeted
The most heretical thing I believe to be absolutely true: the world is actually full of great future founders
Decels believe there is a finite # good startups
We at YC *know with direct experience* it is actually about finding the builder founders who could be great and funding them as early as possible, and giving them a community in which to thrive
David Huang retweeted
Final reminder - last week to apply:
A16Z SPEEDRUN ALPHA is for recent grads and college students who want to start a company but are pre-idea / pre-product
details:
- $20K equity-free upfront to start building
- up to $250K investment when you finalize
- automatic final interview for a16z speedrun, with up to $1M investment
- 8-week, in-person experience with a kickoff retreat, founder AMAs, and small-group dinners alongside the a16z speedrun community
- targeted to early-career highly technical founders
We stayed Alpha bc its a time of great change in the startup community and in the job market. And we know that most of the best founders don’t start with a perfect idea. They start with curiosity, talent, and the willingness to build things until something clicks.
Some of the most important companies of the last decade started this way:
- tinkering with side projects
- hacking on open source
- building weird prototypes with friends
- exploring a space before the opportunity was obvious
The goal of Speedrun Alpha is simple:
find exceptional builders before the idea is fully formed.
Instead of asking you to show up with a polished pitch deck, we give you:
- time
- community
- mentorship
- and just enough capital to start experimenting.
Just show up with technical ambition and curiosity. You spend the summer with us on the a16z Speedrun team exploring ideas, building prototypes, and talking to users. By the end, if something interesting emerges, we help you turn it into a real company.
The kinds of founders we’re looking for tend to look like this:
- engineers who can ship fast
- builders who have shipped side projects before
- hackers who like learning new systems quickly
- people who would probably start companies eventually anyway
We’re intentionally targeting recent grads and college students as a bet on the future.
If that sounds like you — or someone you know — this is the last call.
David Huang retweeted
Building systems that default to Steering instead of Autonomous Agents is the MO for enterprises and personal use cases. AI you truly own (data, weights, execution access). There will always be use cases for autonomous agents but we need significantly more products and discussions in the anti-autonomous agent space.
Replying to @0xSigil
Bro, this is wrong.
Lengthening the feedback distance between humans and AIs is not a good thing for the world.
Today, it means you're generating slop instead of solving useful problems for people. It's not even well-optimized for helping people have fun.
Once AI becomes powerful enough to be truly dangerous, it's maximizing the risk of an irreversible anti-human outcome that even you will deeply regret.
The point of ethereum is to set *us* free, not to create something else that goes off and does some stuff freely while our own situation is unchanged or worsened. (And, as others have pointed out, the models are run by openai and anthropic, so the thing is not even "self-sovereign"; you're actually perpetuating the mentality that centralized trust assumptions can be put in a corner and ignored, the very mentality that ethereum is at war with)
The exponential will happen regardless of what any of us do, that's precisely why this era's primary task is NOT to make the exponential happen even faster, but rather to choose its direction, and avoid collapse into undesirable attractors.
Excited to iterate among philosopher-builders at @cosmos_inst
Introducing our second cohort of AI x Truth-Seeking grantees ↓
Including:
🔹 Ex NotebookLM and DeepMind researchers
🔹 Computer Science professors
🔹 A student "study influencer"
118
David Huang retweeted
Human beings are called to be co-workers in the work of creation, not merely passive consumers of content generated by artificial technology. Our dignity lies in our ability to reflect, choose freely, love unconditionally, and enter into authentic relationships with others. Recognizing and safeguarding what characterizes the human person and guarantees their balanced growth is essential for establishing an adequate framework to manage the consequences of artificial intelligence.
vatican.va/content/leo-xiv/e…
David Huang retweeted
Do the people using the formula "X isn't just Y, it's Z" every other sentence realize that everybody else can see where this is coming from
David Huang retweeted
I’m beyond excited to announce our MIT Press book on Neuroevolution! An HTML version is now available for free on neuroevolutionbook.com/, with a print edition coming out later in 2026.
Real intelligence is not static; it evolves. For decades, the field of neuroevolution has pursued this necessary adaptability. Our book chronicles its development, from early concepts to its modern integration with deep learning and reinforcement learning, exploring its potential for understanding the origins of intelligence and its real-world applications.
And the companion webpage is more than just a book site! It comes equipped with interactive demos, videos, exercises, and tutorials to allow everyone to experience neuroevolution in action. Check it out and let us know what you think!
It was a pleasure to work on this book over the last 4+ years with David (@hardmaru), Yujin (@yujin_tang), and Risto. We are incredibly proud of the result and look forward to celebrating! We hope to connect with many of you at NeurIPS.
We are very grateful to Melanie Mitchell (@MelMitchell1) who provided a fantastic foreword. To quote her: “The next big thing in AI is coming, and I suspect that neuroevolution will be a major part of it”. We think so too!
David Huang retweeted
Who is excited about the @jeffclune show dropping?! Here are some more of my favourite parts including a commentary on AI Safety at the end.


























