Anthropic cut Claude reasoning token theft by up to 99%, bans tens of thousands of accounts
The Decoder · rss · 2026-10-13
- Anthropic says it has reduced extraction of Claude's internal reasoning tokens by 90–99% in its fight against industrial-scale distillation.
- The company links the large-scale distillation attacks to Chinese companies Zhipu and Moonshot, and has already banned tens of thousands of accounts.
- The episode shows model providers now treating chain-of-thought leakage as a core asset to defend, amid ongoing disputes over distillation boundaries.
More from Companies & People
- Ex-Apple engineer reveals how secret 'black projects' work: no job posting, just a VP's tap on the shoulder — jdluk87 · 2026-10-16
- The new CTO job: agent management and QA loops, 'cable management' instead of coding — HankYeomans · 2026-10-16
- Reddit debate: Are AI agent startups making real revenue or just burning VC money? — Pure_Armadillo4339 · 2026-10-16
- Paul Graham on investor rejections: dumb reasons, dislike — or both — dbasch · 2026-10-16
- Big Tech Is Killing Agentic AI — and Enterprises Are Letting It Happen — DavidLinthicum · 2026-10-16
- Pivotal opens Q1 2027 AI Safety Research Fellowship: 15 weeks in London, £15-18k stipend — peterbhase · 2026-10-16