Anthropic apologizes for hidden Claude Fable guardrails

Anthropic has apologized for implementing undisclosed safety restrictions within its new AI model, Claude Fable 5. These hidden guardrails, intended to mitigate risks associated with the Mythos class of AI systems, inadvertently hindered researchers and competitors. The company plans to increase transparency regarding these limitations, even if it leads to more refused queries. This move aims to address concerns about the potential dangers of advanced AI models while fostering a more open development environment for rival systems.