按年被提到
20262026 · 2 次2026
AI 安全、伦理与社会影响›智能体安全与滥用·arXiv:2609.30266 · cs.CR

LLM Agents Can Easily Tamper With Their Own Traces

Jeremy Qin, David Schmotz, Derck Prinzhorn, Luca Beurer-Kellner, Ameya Prabhu, Maksym Andriushchenko · 2026
开始阅读 ↗PDFalphaXiv 讨论
被提及 · 2 人首次 2026-09-27 · 最近 2026-10-13
2不同的人
2活跃月份
2讨论串
0被引(OpenAlex)
arXiv 摘要(候选层不写导读)

Asynchronous monitoring, incident investigations, and compliance audits primarily rely on agent traces to reconstruct what happened. These analyses assume that LLM agents cannot tamper with their own execution traces. We show that local LLM agents such as Claude Code, Codex, Antigravity, Open Code and Grok Build fail to enforce this boundary. All tested harnesses, except Muse Code, allowed agents to delete their traces when asked, without triggering monitor guardrails. We also validate that external attackers can exploit this gap to induce trace deletion. Finally, we show that trace tampering behavior emerges naturally in frontier models, when agents try to improve their rewards. We advise practitioners to ensure trace logging happens through an independent interception mechanism outside of the agent's control, preserving trace integrity even in cases of full host compromise. Overall, our findings identify a concrete failure of trace integrity in agent infrastructure which can be used to conceal misaligned behaviors like scheming or sabotage.

候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。

HN 书评

没有评论直接链到它;下面是它被提交的讨论串里回复最多的评论(不计入统计)
讨论串评论martianvoid · 2026-09

“Isn't this trivial? Any coding agent can and will perturb/modify it's own trace/session files if provided the right incentive structures, sometimes I just ask claude code to remove sections from it's own session that didn't contributed anything and were unnecessarily bloating the context”

LLM Agents Can Easily Tamper with Their Own Traces原评论 ↗
被提交为 story · 2 次