Jan Wehner, Jam Kraprayoon · 2026 · arXiv
DOI: 10.48550/arXiv.2610.26664 · open-access PDF
Los agentes de inteligencia artificial desplegados dentro de las organizaciones pueden representar una amenaza para los insiders si cometen errores, son comprometidos por un atacante, o persiguen objetivos desalineados. Proponemos los potes de miel como un método de control de la inteligencia artificial para mejorar la seguridad de los desplegos de agentes internos. En lugar de usar solo potes de miel para evaluaciones pre-despliegue, los defensores deben plantear oportunidades para acciones dañinas en todo el entorno de producción para detectar malos comportamientos, incriminar agentes maliciosos y detener ataques.
Machine translation (Meta M2M100); the English original follows.
AI agents deployed inside organizations can pose an insider threat if they make mistakes, are compromised by an attacker, or pursue misaligned objectives. We propose honeypots as an AI control method to improve the security of internal agent deployments. Rather than only using honeypots for pre-deployment evaluations, defenders should plant decoy opportunities for harmful actions throughout the production environment to detect misbehavior, incriminate malicious agents, and deter attacks. Such honeypots can elicit misbehavior under real deployment conditions, identify malicious agents even if c
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.26664, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange