Making hardware trojans using AI: artificial intelligence in hardware design and cybersecurity. 📟🤖💉☣️🎰

An international group of academic hardware security experts shared their work last year on how new types of artificial intelligence, especially large language models (LLMs), may help design new microchips and hardware components.

An especially interesting part (for me) is how AI can contribute to hardware security: models and AI-enabled systems can not only design new architectures but also identify weak points and the presence of hardware trojans.

Actually, AI can also identify ideal places and propose the design of hidden malicious circuits - hardware trojans. The paper discuss “test trojans” for debugging purposes, but it’s a matter of how they are used.

Super interesting and very insightful work, full of surprises and ideas for future research. Enjoy the read, and don’t forget to share. Thanks!

More details:

Hardware Design and Security Needs Attention: From Survey to Path Forward [PDF]: https://arxiv.org/abs/2504.08854