Correlated Failures of AI Approval Monitors↗ (원문, 새 창)

본문 미리보기

arXiv:2610.29178v1 Announce Type: new Abstract: Before an AI agent takes a consequential action, such as paying, messaging or changing code, several monitors check it. Their verdicts are combined by aggregating scores or by counting approvals, whatever inputs the monitors share. Monitors that share a base model or a poisoned input, however, may fail together. We borrow the fail-prone systems of distributed computing: lists of monitor sets that may fail together. A rule is safe if it executes no