Matthew Green: machines beat humans at math — why trust human cryptanalysis is complete?
matthew_d_green · x · 2026-10-08
Matthew Green elaborates: we live in a world where machines outperform humans on math problems we fought over for years. Questioning the certainty around human analysis of Module LWE, ECDLP, and already-subexponential factoring, he implies AI may already be making unnoticed progress on the foundations of modern cryptography.
Related event: Cryptographer Matthew Green Warns AI Could Break Public-Key Encryption(4 posts)→
More from Safety
- Streamer to demo prompt-free jailbreak of every open model in live X stream — koltregaskes · 2026-10-08
- Amazon's About You page lets shoppers read AI inferences, including 'has flat buttocks' — aakashgupta · 2026-10-08
- Dev Building Runtime Authorization Layer for AI Agents Seeks Real Teams to Sandbox-Test It — Invisible_act1988 · 2026-10-08
- Devs debate session-level runtime authorization on top of MCP's static config for agents — Longjupping_Tax_3598 · 2026-10-08
- Cryptographer: AI doing amazing math puts elliptic curve hardness in no danger — markjeffrey · 2026-10-08
- ColPali-style visual document indices can be inverted: 47% of words recovered, source page ranked first 98.4% — _reachsumit · 2026-10-08