AI 审批监控器的相关性失效研究

View PDF HTML (experimental)

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 bad action and live if it blocks no good one. We measured twelve small open-weights monitors on payment approval under two pre-registered protocols. Without pre-registration, we also measured three closed frontier models (Claude Opus 5.5, GPT-6 Astra, Gemini 4 Argon) and the open DeepSeek-V3.2. Every reliable fooling of a monitor into approving a fraud came through an evidence source and reached only that source's readers. Fooling it into blocking a correct payment depended on injection resistance, a monitor's refusal to follow instructions planted in its data. Small models and DeepSeek obeyed such notes. The three closed models ignored them in all but one case, and Opus also resisted persuasion tailored to it. On new data, a rule that counts sources instead of votes blocked 0.9% of correct payments, against 7.4% for the best counting rule, at equal safety. The pre-registered worst-case test failed. One attack left the approvers on a single source, so no rule over the twelve language-model monitors survives every measured attack. Adding a rule engine, which cannot be persuaded, restored the condition for a safe and live rule on the measured attacks. Diversify the inputs, harden the monitors, count sources, not votes.

Subjects:	Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC)
Cite as:	arXiv:2610.29178 [cs.CR]
	(or arXiv:2610.29178v1 [cs.CR] for this version)
	https://doi.org/10.48550/arXiv.2610.29178 arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Luca Zanolini [view email]
[v1] Wed, 21 Oct 2026 04:09:37 UTC (73 KB)