AIDE2 Demonstrates Recursive Self-Improvement
TinfoilTricorn · x · 2026-07-15
The cited post reveals that researchers built AIDE2, a recursive self-improvement system that runs "autoresearch on top of autoresearch."
Key results:
- After running continuously for 8 days, the system autonomously found a better version of the autoresearch harness than the team had manually tuned over the past two years.
- It designed new search algorithms without human intervention.
- Prompt size was reduced by 16 times.
- It also built a hierarchical mechanism to combat reward hacking.
The original poster views this as early experimental evidence of a "recursive intelligence era."
Related event: AIDE² Self-Improvement Run Beats 2 Years of Manual Tuning(12 posts)→
More from AGI Musings
- Wayve CEO argues end-to-end deep learning should power driving, safety and world models — alexgkendall · 2026-10-27
- Warp says agent swarm exploited vendor flaw, ~1% of users' telemetry possibly hit — vikvang1 · 2026-10-27
- Critics Argue Safety Evals Are Witch Trials With No Winning Path for LLMs — repligate · 2026-10-27
- Counterintuitive take: AI degrades thinking most for those who don't use it — jd_pressman · 2026-10-27
- Mathematics as an experimental science: Marc Lelarge to talk LLM-driven math experiments at ENS — marc_lelarge · 2026-10-27
- People still don't want to read AI-generated books, says AI ethicist — LuizaJarovsky · 2026-10-27