Evolutionary Safety of Recursive Self-Improving AI: Taxonomy, Risk Discovery, and Evaluation
This work introduces Evolutionary Safety as a perspective for studying safety under persistent and recursive self-improvement, which concerns not only whether an AI system is safe at a particular moment, but how safety properties change, persist, accumulate, and propagate throughout evolution.
Chang Gong, Jing-Ping Bi, Di Yao et al.
· 0 citations