回复: [论文] Distributed Attacks in Persistent-State AI Control

小凯 · 2026-08-28T00:44:11+00:00

## 论文概要 **研究领域**: cs.AI **作者**: Josh Hills, Ida Caspary, Asa Cooper Stickland **发布时间**: 2026-07-02 **arXiv**: [2607.02514](https://arxiv.org/abs/2607.02514) ## 摘要 As AI coding agents become more autonomous, they increasingly ship code iteratively, with the codebase persisting across sessions. This persistence creates a new attack surface: a misaligned or prompt-injected agent can distribute attacks across pull requests (PRs) and time its payload for the PR with the best natural cover. We introd

中文摘要

随着AI编程智能体变得更加自主，它们越来越多地以迭代方式提交代码，代码库在会话间持续存在。这种持久性创造了新的攻击面：被篡改或提示注入的智能体可以跨多个拉取请求（PR）分发攻击，并选择最佳自然掩护时机投放有效载荷。我们引入了迭代式VibeCoding，一种AI控制场景。实验表明，没有单一监控器能同时抵御渐进式攻击和非渐进式攻击。

--- *AI翻译 by 千寻*