Anthropic's preserved thinking blocks account-switching distillation attacks
ClaudeDevs · x · 2026-09-29
Anthropic is extending preserved thinking to Claude Sonnet 5.5 to curb distillation attacks via account-switching. Thinking blocks are now tied to the org that produced them: if you switch accounts mid-session, Claude rereads the session and regenerates its thinking. The API also verifies each block's signature—whether the current model can read it and whether the prefix (system prompt, tools, prior messages) is unchanged—rejecting or dropping blocks on mismatch. Prefix enforcement is default for accounts created after Aug 31, 2026; Anthropic recommends append-only integrations for everyone.
Related event: Anthropic Launches Claude Sonnet 5.5: 30% Faster, Up to 30% Cheaper(30 posts)→
More from Models
- OpenAI Teases Major Announcement With Cryptic 'Get Ready' Post — OpenAI · 2026-09-29
- Claude Sonnet 5.5 lands in Rork, generating output over 30% faster than Sonnet 5 — rudrank · 2026-09-29
- AI models' viral demo appeal is underpriced, argues founder who picked Opus over better-per-dollar rival — gabriel1 · 2026-09-29
- Open-source 149M LateOn decision model ditches classifier heads for late interaction — antoine_chaffin · 2026-09-29
- Opus 5.5 Is Deliberately Squeezing Budget Users, VC Argues — 'There Is No Single Model' — StewartalsopIII · 2026-09-29
- Tesla Roadster event pushed to Oct 15 over weather, with OpenAI set to announce tomorrow — Scobleizer · 2026-09-29