Anthropic apologizes for invisible Claude Fable guardrails

Anthropic admitted that it stealthily throttled its new Claude Fable 5 model with hidden guardrails that hinder researchers and rivals from using it. The firm says it will remove those restrictions and be more transparent about when they trigger, even if it means the model refuses more queries.

Summary written by Kernelia from the original article by The Verge — AI. The story and its rights belong to its author.