Cryptographer Matthew Green pushes back on LLM cryptanalysis fears: best AES-128 attacks still need 2^89 work
matthew_d_green · x · 2026-11-03
Johns Hopkins cryptographer Matthew Green debated whether LLMs pose a real threat to symmetric cryptography, arguing that claims of symmetric crypto being more vulnerable than public-key crypto lack any real support — likening them to expecting a polynomial-time 3SAT algorithm next month.
He notes the state of the art: the best known attacks on AES-128 only reach 7 rounds (requiring 2^89 work and 2^105 queries), with ChaCha20-256 at 7.5/20 rounds — far below deployed security frontiers, making LLM-assisted attacks a non-issue for now.
Related event: Cryptographers Debate Whether LLMs Could Break Symmetric Encryption(6 posts)→
More from AGI Musings
- Michael Levin 用「计算器 slop」类比回击 AI slop 之争 — debreuil · 2026-11-03
- Labs' underinvestment in risk communication is a large-scale ethical mistake, argues researcher — harris_edouard · 2026-11-03
- Engineers Say AI Makes Them 3x Faster but Rate Their Work Only 1.4x-2x More Valuable — sanjaykalra · 2026-11-03
- Webflow AI revenue GM: redesign your org as if you had 1/8 of today's headcount, with agents in mind — akumar · 2026-11-03
- Edouard Harris: radically cheaper cyber attacks mean future infrastructure disruptions trace back to AI — harris_edouard · 2026-11-03
- Academic complains conference talks all feel the same as AI homogenizes slide decks — CSProfKGD · 2026-11-03