LLMPirate: LLMs for Black-box Hardware IP Piracy
Vasudev Gohil
Network and Distributed System Security (NDSS) Symposium 2025 · Day 2 · Hard- & Firmware Security
In the rapidly evolving landscape of Large Language Models (LLMs), their applications span numerous domains, from finance and education to software development and even semiconductor chip design. While LLMs offer immense promise, their capabilities also introduce new vectors for sophisticated attacks. This talk by Vasudev Gohil, presented at the NDSS Symposium, delves into one such novel attack: **LLMPirate**, a technique that leverages LLMs to facilitate **black-box hardware intellectual property (IP) piracy**. The research explores how generative AI can be weaponized to structurally alter hardware circuits at the gate level, rendering them undetectable by existing piracy detection tools while preserving their original functionality.
AI review
Genuinely novel intersection of LLM capability and hardware security — using generative AI to perform black-box adversarial evasion of gate-level IP piracy detectors is a contribution the field hasn't seen before. The methodology is rigorous: prompt syntax translation, divide-and-conquer decomposition, and formal equivalence verification in a feedback loop are real engineering solutions to real LLM limitations, not hand-waving. Docks one star because the defensive analysis is shallow and the threat model has practical limits the talk undersells.