Recursive Self-Improvement in AI: From Bounded Self-Refinement to Autonomous Research Loops
Updated
Updated · arxiv.org · Jul 8
Recursive Self-Improvement in AI: From Bounded Self-Refinement to Autonomous Research Loops
1 articles · Updated · arxiv.org · Jul 8
Summary
A comprehensive survey of 1,250 recent AI papers maps the state of recursive self-improvement (RSI) in artificial intelligence.
The study organizes advancements by what AI systems improve and how much human oversight remains, highlighting evaluator reliability as the key constraint.
While bounded self-refinement is now industrial practice, fully autonomous RSI is limited by verification, compute, and the need for human judgment.