It separates true recursive self-improvement from using AI to automate parts of AI development, and explains why verifying gains is hard because of variance and reward hacking. Useful for judging self-improving AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition → claims.
Terms in this piece · Glossary
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
