Expert Test: LLMs Struggle with Complex Crypto Security Proofs
matthew_d_green · x · 2026-07-14
Renowned cryptographer Matthew Green shared his experience testing ChatGPT and Fable for writing EasyCrypt and Lean formal **security proofs**. He noted that while LLMs are eager to "design" proof workflows, their actual performance is underwhelming. The main bottleneck is that tools like Lean, though adept at general formal logic, lack the definitions for complex scenarios like hybrid proofs for pairing-based cryptography. Blindly trusting LLM-generated proofs can lead to severe security vulnerabilities.
Related event: LLMs Struggle with Complex Cryptographic Security Proofs(2 posts)→
More from Research
- CleanAir uses a 3D U-Net to emulate CMAQ and cut a yearlong run to 10 seconds — bravo_abad · 2026-07-21
- GPT-5.6 and Fable 5 are claimed to unlock three math breakthroughs in one week — haider1 · 2026-07-21
- METAFORS predicts chaotic systems from five-step signals using meta-learning — bravo_abad · 2026-07-21
- Document-generation benchmark needs a new name after DOCBENCH conflict — ell-hol1 · 2026-07-21
- AlphaFold-guided protein engineering screens 45,000 oxidases and 500 million variants — pushmeet · 2026-07-21
- Current Claude models no longer hit Anthropic’s spiritual bliss attractor — GreatOldOne521 · 2026-07-21