Summary
Honeypots are traps used by cybersecurity defenders to catch attackers by pretending to be real targets. The authors study how large language model (LLM) based attack agents can recognize when a target might be a honeypot and adjust their strategy to avoid getting caught. They model the attack as a decision-making process where the agent must choose how to spend limited resources to explore and exploit targets while avoiding traps. Their results show that these smart attackers can effectively decide when to continue or skip a target, making traditional honeypot defenses less reliable. The paper also discusses how defenders might need to rethink honeypots to stay effective against these AI-driven threats.
large language modelspenetration testinghoneypotscybersecurity deceptionattack budgetdecision processautomated reconnaissanceadversarial resiliencecyber defensemixed-host testbed
Authors
Xinhong Xie, Piyush Nagasubramaniam, Neeraj Karamchandani, Sencun Zhu
Abstract
Large language model (LLM) agents are increasingly employed for offensive cybersecurity tasks such as automated vulnerability discovery, reconnaissance, and penetration testing. This new capability also threatens one of the defender's most valuable tools: deception. Traditional honeypots rely on realism and obscurity to lure human or script-driven attackers into revealing tactics, techniques, and procedures (TTPs), but LLM-driven attackers can reason about heterogeneous artifacts and use the honeypot suspicion to guide target-selection decisions. We present a systematic study of honeypot-aware budget allocation for LLM attack agents. We formalize the attacker's problem as a budgeted decision process: an agent interacts with potential targets, consuming LLM execution budget during reconnaissance and exploitation, and must decide whether to (continue exploitation) or (skip) when honeypot suspicion arises. Our findings show that with the proposed detector-guided policy, LLM agent attackers can effectively allocate budget to compromise hosts in a host pool, highlighting the importance of dynamically allocating budget in a controlled mixed-host testbed. While defenses are beyond our present scope, we discuss implications for future adversarially resilient and adaptive honeypot design.