Recursive Self-Improvement in AI: From Bounded Self-Refinement to Autonomous Research Loops

A comprehensive survey of 1,250 arXiv papers (2024-2026) on AI self-improvement reveals a taxonomy distinguishing bounded self-refinement from open-ended recursive self-improvement (RSI), with the latter constrained by grounding, collapse dynamics, and compute limits. The study identifies a verification hierarchy for self-evaluation signals and warns that failure modes like self-confirming loops and model collapse follow from violations of this hierarchy, while the human-in-the-loop bottleneck persists at the top of the hierarchy. The authors call for governance-grade measurement of self-improvement as a critical unmet need.

arXiv:2607.07663v1 Announce Type: new Abstract: AI systems increasingly participate in their own improvement: revising their outputs, adapting their own harnesses during deployment, training on data they generate, and, increasingly, conducting AI research itself. This literature is described under a vocabulary "self-refine," "self-reward," "self-play," "self-evolve" that conflates fundamentally different ambitions. We survey 1,250 arXiv papers 2024-2026 along two axes: what the system improves -- its behavior in deployment, its policy through training, its evaluator, or the research process itself -- and the degree of loop closure human-in-the-loop to fully closed . The taxonomy separates bounded self-refinement -- convergent, evaluable, and already industrial practice -- from open-ended recursive self-improvement RSI , which remains bounded by grounding requirements, collapse dynamics, and compute constraints on every measured axis. Its distinctive feature is a dedicated category for self-evaluation: every improvement loop is a claim that some signal can substitute for human judgment. We survey the evaluator design space -- judges, process reward models, verifiers, rubrics, meta-evaluation -- order the signals into a verification hierarchy from formal verifiers strongest to intrinsic self-assessment weakest , and observe that demonstrated self-improvement strength tracks this hierarchy, that its failure modes self-confirming loops, model collapse, diversity collapse follow from its violations, and that the "research direction-setting" bottleneck keeping humans in the loop sits at the top of that hierarchy. We connect the technical literature to the theory of RSI limits and to the safety and governance questions raised by frontier-lab accounts of closing the loop, and identify governance-grade measurement of self-improvement as the field's most underpopulated niche.