arXiv 论文证明 AI 模型可借助密码学后门隐藏任意能力

Hacker News：AI 热帖· sbulaev·2026-10-13 10:07· 3 小时前AI 评分61

AI Models Can Provably Hide Arbitrary Capabilities

AI 导读

arXiv:2610.13928 论文构造后门攻击，把加密神经网络嵌入宿主模型权重，仅在收到秘密触发器时解密执行，在密码学假设下即使完全访问权重也难以诱导和解释。

正文

Abstract:Capability evaluations aim to surface dangerous model behaviors before deployment, but their reliability depends on the assumption that hidden capabilities can be elicited. We show this assumption does not hold under adversarial conditions by constructing backdoor attacks that embed encrypted neural networks within host model weights, decrypting and executing them only upon receiving a secret trigger. Using digital lockers, these hidden circuits remain provably hard to elicit and interpret under cryptographic assumptions, even given full access to model weights, extending prior work from hiding fixed strings to arbitrary computations. We implement our proof of concept constructions in PyTorch and evaluate elicitation through fine-tuning, weight noising, gradient-based trigger search, and latent adversarial training on a small chemical reaction prediction task - a proxy for CBRN-relevant capabilities. Our results demonstrate resistance to these methods and, together with our cryptographic guarantees, reveal a limit of elicitation-based capability audits. We release our models to support the improvement of defenses.

Submission history

From: Andis Draguns [view email]
[v1] Fri, 9 Oct 2026 17:55:06 UTC (878 KB)

来源：Hacker News：AI 热帖 · arxiv.org