Encrypted Reasoning Traces Are Portable Across Sessions, Research Confirms; Mitsuhiko Calls for Visibility
mitsuhiko · x · 2026-08-11
Armin Ronacher (mitsuhiko) highlighted a security research report on AI models' encrypted reasoning traces, arguing that since labs confirm no security issues with revealing them, they should be made visible to developers.
Background context notes that in May, research found encrypted reasoning could be replayed outside its original context. The new report further confirms that these encrypted thoughts are fully portable across sessions, users, and even different models within the same provider, though labs maintain this poses no security implications.
More from Models
- Devs say LLM coding mistakes are a different class: overcomplication, inferred limits, worse at higher reasoning — Mu5_ · 2026-10-16
- ChatGPT UI screenshot reveals Wallet coming soon plus Email, Calendar, Health and Finances integrations — jxnlco · 2026-10-16
- Why the next OpenAI and Anthropic models may still disappoint: no native audio or video input — imjustnewatai · 2026-10-16
- Thousands Hit by ChatGPT Standard Voice Mode Regression: Wrong Voice, Interruptions — shaunralston · 2026-10-16
- Gemini Argon 4 Quota Exhausted, Reset Countdown Shows Nearly 7 Days — ssh4net · 2026-10-16
- AINews: Meta ships 816B Muse Spark Open; Goodfire argues neural nets are already clean — Latent Space · 2026-10-16