{"id":242,"date":"2026-10-13T10:09:01","date_gmt":"2026-10-13T10:09:01","guid":{"rendered":"https:\/\/scoy.ai\/guides\/anthropic-distillation-defenses-bill-blocked-requests\/"},"modified":"2026-10-20T08:02:02","modified_gmt":"2026-10-20T08:02:02","slug":"anthropic-distillation-defenses-bill-blocked-requests","status":"publish","type":"post","link":"https:\/\/scoy.ai\/guides\/anthropic-distillation-defenses-bill-blocked-requests\/","title":{"rendered":"AI News Roundup for October 13: Anthropic Says It Cut Claude Distillation by 90 to 99 Percent, and It Bills You for Blocked Requests"},"content":{"rendered":"\n

Anthropic said Monday that its new defenses cut the reasoning data harvested by distillers by around 90 to 99%<\/a>. The coverage ran that number and the two Chinese labs the post names, Zhipu and Moonshot.<\/p>\n\n\n\n

It is Anthropic grading its own homework, measured “on our own platform.” The same post says distillers also route traffic through third-party clouds and other inference providers.<\/p>\n\n\n\n

The bill sits near the bottom. Legitimate developers had to “update their integrations,” the classifier wrongly blocks “fewer than 0.1%” of harmless requests, and “we now charge users for those blocked requests.”<\/p>\n\n\n\n

Breaks your stack.<\/strong> The mechanism is called preserved thinking. On Fable 5.1 and Opus 5.5, an API account created on or after August 31 cannot change anything that came before a thinking block it sends back.<\/p>\n\n\n\n

Anthropic’s docs<\/a> define “anything” as the system prompt, the tool list and every earlier message. Stamp a fresh date into the system prompt mid-session, send the earlier thinking back, and the request dies with a 400.<\/p>\n\n\n\n

Fewer than 0.1% sounds like nothing. By my arithmetic, at a million requests it is up to a thousand you paid for and got nothing back.<\/p>\n\n\n\n

Matters.<\/strong> OpenAI raised the default context for GPT-6.1 Sol in Codex and Work to 500k tokens<\/a>, up from 272k. It stays configurable up to 1M.<\/p>\n\n\n\n

My read: fewer compactions, and a long session draws more from your plan on every turn. Pick the number on purpose.<\/p>\n\n\n\n

Marketing.<\/strong> Meta shipped Muse Wonder on Monday. In the Muse app it needs a $20 or $100 monthly plan<\/a>, and the larger Muse Wonder Full is limited to Muse Code subscribers at $50 or $200.<\/p>\n\n\n\n

Meta AI chief Alexandr Wang told CNBC the API version will be priced “meaningfully below” OpenAI and Anthropic. That is an adjective, and I budget against numbers. Meta priced its whole app portfolio last month<\/a>, so it knows how to publish one.<\/p>\n\n\n\n

\n