Anthropic's Distillation Defense Claim Draws Public Skepticism

Episode 1 · Anthropic Says Anti-Distillation Defenses Cut Data Theft by 90-99%, Names Zhipu and Moonshot (2026-10-13, 20 posts)

On October 13, Anthropic published an article on "distillation defenses," revealing that over the past few months its new defensive methods reduced the amount of data stolen from Claude via distillation attacks by roughly 90%-99%, and that it turned away substantial revenue and paid traffic totaling trillions of tokens to do so.

Confirmed

• Anthropic officially stated that the new anti-distillation methods have cut the data harvested by distillers by about 90%-99% (m1, m3, and m4 all relay the same official figure)
• To implement this defense, Anthropic declined substantial revenue and paid requests totaling trillions of tokens (m4)
• The article called out Zhipu by name: during the development of GLM-5.3, it distilled cyber-offense-and-defense capabilities from leading US models, attempted to distill Claude Fable 5, and took follow-up actions after discovering that Anthropic's cybersecurity guardrails weakened attack effectiveness (m2, m3)
• One distiller's network controlled over 20,000 fake accounts (a specific figure within the hundreds-of-thousands range: "over 20,000"), covertly extracting Claude's reasoning outputs at scale (m5)
• Design rationale of the defense: ordinary users rarely ask a model to output full hidden reasoning, so suspected distillers' extraction requests can be detected this way; the initial guardrail release was bypassed, after which distillers scaled up (m5)

Why it matters

• Distillation attacks use vast numbers of fraudulent accounts to covertly extract leading models' capabilities at industrial scale and cheaply replicate frontier models; the effectiveness of anti-distillation directly bears on frontier labs' moats
• @abhiadesai commented that this is exactly why open-source model quality is a "lagging indicator"—open-source models' quality typically lags behind the closed-source frontier models they can distill from

• Anthropic says new defenses cut distillation attack data harvesting by 90-99%, cites Zhipu's GLM-5.3 — AnthropicAI · 2026-10-13
• Anthropic says it cut distiller data harvesting by 90-99%, turning away trillions of tokens — abhiadesai · 2026-10-13
• Anthropic Says Zhipu Distilled Claude to Train GLM-5.3, Calls It Top Cyber Open-Weights Model — No-Point1424 · 2026-10-13
• Anthropic says new defenses cut distillation-harvested data by 90-99%, names Zhipu's GLM-5.3 — scaling01 · 2026-10-13
• Leaked Anthropic anti-distillation details: one network ran 20,000+ fake Claude accounts — teortaxesTex · 2026-10-13
• Anthropic cuts distilled data harvesting by 90–99%; David Sacks says no regulation needed — DavidSacks · 2026-10-13
• Anthropic's anti-distillation doc doubles as a guide to avoiding wrongful Claude bans — op7418 · 2026-10-13
• Anthropic says it has significantly blunted distillation attacks against Claude — AlexTensor · 2026-10-13
• Anthropic's anti-distillation defenses may have cut data harvest by 90-99% — Proper_Actuary2907 · 2026-10-13
• Anthropic says new safeguards cut Claude distillation by 90-99%, naming Zhipu and Moonshot — kimmonismus · 2026-10-13
• Anthropic details anti-distillation wins: data leakage down 90-99%, accuses Zhipu's GLM-5.3 — kimmonismus · 2026-10-13
• Anthropic says it blocked 90-99% of rival attempts to harvest Claude reasoning for distillation — mark_k · 2026-10-13
• Anthropic cuts distillation data leakage by 90-99%, bans tens of thousands of accounts — 新智元 · 2026-10-13
• Anthropic cuts distillation attack exposure 65x with new Claude safeguards — CurieuxExplorer · 2026-10-13
• Anthropic says it sharply curbed distillation attacks on Claude, drawing hypocrisy criticism — TheMoonMidas · 2026-10-14
• Anthropic says it cut data harvestable from Claude via distillation attacks by 90-99% — evisoft · 2026-10-14
• Anthropic details cross-session replay attacks used to steal Claude's chain-of-thought — Miles_Brundage · 2026-10-14
• Anthropic says new defenses cut distillation-attack data theft by 90–99%, names Zhipu — pstAsiatech · 2026-10-14
• Anthropic says it has blunted distillation attacks on Claude; researcher pushes back on the term — AlexTensor · 2026-10-14
• Anthropic says it blunted distillation attacks on Claude; skeptics poke at data hypocrisy — AlexTensor · 2026-10-14

Episode 2 · Anthropic's Distillation Defenses Draw Public Criticism (2026-10-14, 2 posts)

Anthropic announced strengthened defenses against distillation attacks, but developer fivefilters criticized the company for publicly bragging about hiding model reasoning chains in the name of protection.

• Blogger blasts Anthropic for boasting about hiding models' reasoning chains — fivefilters · 2026-10-14
• Anthropic boosts distillation attack defenses; Reddit post questions its 'reasoning protection' stance — fivefilters · 2026-10-14