recursive self-improvement

Google and DeepMind researchers introduced Dream-RSI, a method that lets AI agents test new search strategies by replaying recorded past attempts instead of running costly new computations. Tested on Gemini 3.1 Pro and Gemini 3.7 Flash across eight tasks, it matched or beat baselines while using far fewer attempts.

OpenAI's GPT-6 Astra took first place on ulam.ai's ErdosBench, solving 106 of 226 open math problems and disproving 27 others. Chief scientist Jakub Pachocki says the company deliberately didn't optimize for math research, prioritizing recursive self-improvement work instead.

A new Anthropic paper from its fellows program shows an automated AI system improving performance on all 10 tested alignment benchmarks, outperforming experienced human researchers within six hours at a fraction of the cost. The research, led by Anthropic Fellow Chen Yueh-Han, is described as early evidence that automated alignment post-training could become practical soon.

IAPS fellow Severin Field interviewed 25 researchers from top AI labs about recursive self-improvement in late 2025. Several milestones they cited as evidence of progress — Math Olympiad gold, autonomous training loops, majority AI-written code — have since occurred, according to a new report.