AI-Assisted Design of a Post-Quantum Cryptographic Accelerator
This study presents an AI-assisted accelerator design for post-quantum cryptography, addressing trust and authorship issues.
The transition to post-quantum cryptography is mandated by published timelines. Defective silicon cannot be patched remotely, and standard acceptance gates fail to detect certain classes of ML-DSA defects. The developed accelerator successfully passed known-answer tests (KAT) while implementing a norm check, revealing that the final coefficients of candidates were left unverified. This research highlights significant findings regarding the trust and authority of AI authorship in cryptographic artifacts.
This synthesis was produced from its source by AI; there is no human editor or manual review step. How we work