Our progress toward recursive self-improvement, and its implications.
Import AI 455: AI systems are about to start building themselves.
The first step towards recursive self improvement
Harness Engineering for Self-Improvement
The concept of recursive self-improvement (RSI) dates back to I. J. Good (1965), where he defined an “ultraintelligent machine” as a system that can surpass humans in all intellectual activities and design better machines to improve itself. Yudkowsky (2008) used the phrase “recursive self-improvement” for a specific feedback loop: an AI uses its current intelligence to improve the cognitive machinery that produces its intelligence. This feedback loop in modern AI may indicate the model rewriting its own weights directly, or more broadly the model improves the training pipeline and the deployment system, which in turn enables a better successor model with improved performance across economically valuable tasks. The speed of research development in AI has been shown to drastically accelerated in frontier labs (Anthropic; OpenAI).
Peter Wildeford🇺🇸🚀 on Twitter / X
Great to see @tristanharris talking on his podcast about recursive self-improvement. Here's how his guest, Tim Fist, puts it -- "Over the last few months we've had all three of the leading US AI labs say that having the option for a global slowdown or pause in AI development is… https://t.co/s5UURtkEjQ— Peter Wildeford🇺🇸🚀 (@peterwildeford) June 21, 2026
RSI Simulator: Play the Economics of Recursive Self-Improvement
Bootstrap an AI lab from scratch. A web game from Paradigm built on real economic models of AI recursive self-improvement.
AI Research Evaluation: Negative Findings and Failure Modes | Arvind Narayanan posted on the topic | LinkedIn
📢AI agents can autonomously conduct AI research when the result is easily verifiable, but what about open-ended AI research? That’s much harder to study, and our new preprint is our first crack at doing so. Our main finding is negative, and we identify five recurring failure modes. https://lnkd.in/eGKYi4Sa Our results are tentative, and we are working to address the limitations (sample size, potential scaffold improvements). But if the finding holds up, what are the implications? It depends on whether you think recursive self improvement can be achieved simply by hill climbing at scale (I personally don’t think so) and whether you think current limitations of open-ended research like judgment and creativity could change quickly (I’m personally very open to this possibility). We plan to continue this style of evaluation — which we call shadow evaluation — on a regular basis. We’ve wanted to do this for two years, but it took so long because we wanted to get the method right. The idea behind shadow evaluation was suggested by some of the UK AISI coauthors of the paper and refined by the Princeton team. This method has important advantages (and limitations) over the current ways of evaluating agents’ ability to conduct AI research. If you’re an AI researcher interested in working with us on a shadow evaluation based on one of your papers, we’d love to hear from you. https://lnkd.in/ecyp55SW This type of evaluation necessarily involves a ton of researcher flexibility in design, execution, and interpretation. Members of the core team have a particular position in the debate on recursive self-improvement / superintelligence, and this could influence how we conduct the research. We have a detailed section in the paper on our potential biases and how we address them. We sought out a team of collaborators who don’t all share our priors, and we explicitly surface the interpretive disagreements that resulted. For future evaluations, we are interested in having “adversarial collaborators” as part of the core team. This paper exists because of the careful, time-consuming and very much human work that Peter Kirgis, Sayash Kapoor, Andrew Schwartz, and Stephan Rabanser did over the last few months. I’m also very grateful to the larger group of collaborators and co-authors. The work is part of the larger CRUX project that pushes frontier AI agents beyond what benchmarks can measure (https://cruxevals.com/). We are looking for a senior researcher to join the team: https://lnkd.in/e9dC22X5
Why I work on self-improving AI despite the risks - Jeff Clune
Why I work on self-improving AI despite the risks. Jeff Clune.
Are we offloading too much of our thinking to AI?
Reflections on autonomy and the value of thinking for ourselves
Itai Yanai on Twitter / X
Unpopular opinion: Going straight to AI limits your creatively, because it short-circuits the iterative process you need to develop new ideas. pic.twitter.com/O2iPs0bfh0— Itai Yanai (@ItaiYanai) June 20, 2026
The Jevons Paradox of AI - Wesley's notes
Why AI can make us more productive but will never save us time
Jenny Zhang on Twitter / X
Introducing Hyperagents: an AI system that not only improves at solving tasks, but also improves how it improves itself.The Darwin Gödel Machine (DGM) demonstrated that open-ended self-improvement is possible by iteratively generating and evaluating improved agents, yet it… pic.twitter.com/YJPFTJ51SO— Jenny Zhang (@jennyzhangzt) March 23, 2026
Think First, AI Second
Three principles for keeping your cognitive edge while leveraging AI's capabilities
We build AI that works for humans
Imbue builds AI to help people think, create, and build. We share our tools openly because we believe progress in AI should be collaborative and developer-driven
Thinking Fast, Slow, Artificially: AI and Your Brain
AI can boost your decisions — or quietly undermine them. Wharton researchers introduce a new theory of cognition that every leader needs to understand.
What happens when AI replaces the parts of work people love?
We’re using AI to move faster. I’m not yet convinced we’re using it to work better.
