How Intervention History Changes AI Agent Sabotage on Subsequent Tasks↗ (원문, 새 창)

본문 미리보기

arXiv:2610.26935v1 Announce Type: new Abstract: AI control evaluations often reset agents' memory between tasks, although deployed agents may retain records of prior interventions. We ask whether learning that an earlier attack was caught changes an agent's ability to evade defer-to-trusted (DtT), a protocol that replaces suspicious work with a trusted model's solution. Herein, we study cross-episode adaptation: whether intervention history changes attack behavior on a subsequent task with diff