{"slug": "fable-5-2-reportedly-injects-a-hidden-copyright-reminder-then-blocks-the-chat-it", "title": "Fable 5.2 reportedly injects a hidden copyright reminder, then blocks the chat inspecting it", "summary": "Anthropic's Claude Fable 5.2 repeatedly injected a hidden copyright reminder into an API user's conversation after the words \"pirated\" and \"torrent\" appeared, then blocked the chat when the user tried to inspect the reminder, according to an account posted on X by a user identified as Jake (@v01dpr1mr0s3). Jake reported that the session was consumed by an instruction he never wrote and could not see in the normal interface, and that he could remove it only by rewinding the conversation; Anthropic documents both the classifier reminders and the HTTP 200 refusal format, leaving open how often a benign trigger word keeps an ip_reminder firing turn after turn and how developers should debug a refusal labeled as a success.", "body_md": "# Fable 5.2 reportedly injects a hidden copyright reminder, then blocks the chat inspecting it\n\n**Jake says two words, \"pirated\" and \"torrent,\" triggered a copyright reminder Fable 5.2 could not stop discussing, then a block when he inspected it.**\n\n        By [Ryan Merket](https://runtimewire.com/author/ryan-merket)\n        · Published \n\nPrimary source: [x.com](https://x.com/v01dpr1mr0s3/status/2119993017717358913)\n\n## Why it matters\n\nAPI customers pay for every token Fable 5.2 spends reasoning, and Jake's account describes a session consumed by an instruction the customer never wrote, could not see in the normal interface, and could not remove except by rewinding the conversation. Anthropic documents both the classifier reminders and the HTTP 200 refusal format, leaving questions about how often a benign trigger word keeps an ip_reminder firing turn after turn, and how a developer is supposed to debug a refusal that arrives labeled as a success.\n\nAn [API user who posts as Jake (@v01dpr1mr0s3)](https://x.com/v01dpr1mr0s3) says Anthropic's [Claude Fable 5.2](https://runtimewire.com/models/anthropic/claude-fable-5.2) repeatedly inserted a hidden copyright reminder into his conversation, fixated on the insertion, and then blocked a request on reverse-engineering grounds when he tried to inspect it.…\n\nWhen you purchase through links in our articles, [we may earn a small commission](https://runtimewire.com/affiliate-monetization-standards). This doesn’t affect our editorial independence.", "url": "https://wpnews.pro/news/fable-5-2-reportedly-injects-a-hidden-copyright-reminder-then-blocks-the-chat-it", "canonical_source": "https://runtimewire.com/article/fable-5-2-hidden-ip-reminder-reverse-engineering-block", "published_at": "2026-11-10 18:57:06+00:00", "updated_at": "2026-11-10 22:21:13.018310+00:00", "lang": "en", "topics": ["large-language-models", "ai-safety", "ai-products", "artificial-intelligence"], "entities": ["Anthropic", "Claude Fable 5.2", "Jake", "@v01dpr1mr0s3", "Ryan Merket"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/fable-5-2-reportedly-injects-a-hidden-copyright-reminder-then-blocks-the-chat-it", "markdown": "https://wpnews.pro/news/fable-5-2-reportedly-injects-a-hidden-copyright-reminder-then-blocks-the-chat-it.md", "text": "https://wpnews.pro/news/fable-5-2-reportedly-injects-a-hidden-copyright-reminder-then-blocks-the-chat-it.txt", "jsonld": "https://wpnews.pro/news/fable-5-2-reportedly-injects-a-hidden-copyright-reminder-then-blocks-the-chat-it.jsonld"}}