Anthropic's Latest AI Models Run Longer on Their Own—But Safeguards Lag Behind

Anthropic released Claude Fable 5 and Mythos 5 in June 2026, both able to work autonomously for extended periods without human oversight. The longer these systems run unsupervised, the further problems can spread before someone catches them. The pace of capability development is outrunning the safeguards and oversight systems designed to manage them—a familiar gap in technology history.