Wired: Anthropic Walks Back Policy That Could Have 'Sabotaged' AI Researchers Using Claude Fable 5

Wired reports Anthropic has responded to — and per the Reddit post title, walked back — the Claude Fable 5 policy that silently throttled model performance for frontier AI researchers without user notification. The original policy, buried in the 319-page system card, used prompt modification, steering vectors, or PEFT to limit effectiveness on LLM pretraining and infrastructure tasks without telling users. Specific changes Anthropic announced could not be verified as the article was inaccessible; the backlash had drawn sharp criticism from Nathan Lambert (AI2), Jeremy Howard (Fast AI), and the Hugging Face team.