Honeypot Protocol

Computer Science > Cryptography and Security [Submitted on 14 Apr 2026] Honeypot Protocol Najmul Hasan Trusted monitoring, the standard defense in AI control, is vulnerable to adaptive attacks, collusion, and strategic attack selection. All of these exploit the fact that monitoring is passive: it observes model behavior but never probes whether the model would behave differently under different perceived conditions. We introduce the honeypot protocol, which tests for context-dependent behavior by varying only the system prompt across three conditions (evaluation, synthetic deployment, explicit no-monitoring) while holding the task, environment, and scoring identical. We evaluate Claude Opus 4.6 in BashArena across all three conditions in both honest and attack modes. The model achieved 100% main task success and triggered zero side tasks uniformly across conditions, providing a baseline for future comparisons with stronger attack policies and additional models. Comments: 8 pages, 1 figure, 1 table. Research conducted at the AI Control Hackathon, March 2026. Code: this https URL Subjects: Cryptography and Security (cs.CR) Cite as: arXiv:2604.13301 [cs.CR]   (or arXiv:2604.13301v1 [cs.CR] for this version)   https://doi.org/10.48550/arXiv.2604.13301 Focus to learn more Submission history From: Najmul Hasan [view email] [v1] Tue, 14 Apr 2026 21:13:54 UTC (16 KB) Access Paper: HTML (experimental) view license Current browse context: cs.CR < prev   |   next > new | recent | 2026-04 Change to browse by: cs References & Citations NASA ADS Google Scholar Semantic Scholar Export BibTeX Citation Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Demos Related Papers About arXivLabs Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)