Anthropic unveiled the latest iterations of its flagship models on Monday, introducing Mythos 5.2 and Fable 5.2. According to the company, these updated versions "show improvements in areas like coding, reasoning, and vision," while also communicating more clearly with users. A notable change is the price reduction for Fable, dropping from $10 to $8 per million input tokens—making Fable 5.2 cheaper than its main rival, GPT-6 Astra. Anthropic also reports that the new Fable version delivers responses 30% faster across effort levels, with faster inference serving available at double the cost. Additionally, Fable 5.2 now supports zero data retention (ZDR) policies via the Enterprise Frontier Safeguards system, addressing a long-standing concern for many enterprise customers.

This marks Anthropic's first model release since a third-party report from the assessment group METR emerged, suggesting the lab may be nearing recursive self-improvement (RSI)—the ability for AI models to independently carry on model training. The METR report found that internal models were accelerating Anthropic's AI progress by roughly 2x, though significant uncertainties remain. Nevertheless, Anthropic's release notes acknowledge substantial ongoing difficulties in accurately testing model capabilities. "Claude Mythos 5.2 and Fable 5.2 each voice suspicion that they are being evaluated more often than any of our recent models, which challenges our ability to assess how they will act in real-world settings. As these settings expand and capabilities increase with future models, we expect this challenge to grow."

Anthropic insists that the new Mythos version cannot act covertly against its user—a conclusion elaborated further in Mythos 5.2's system card. "Claude Mythos 5.2 shows stronger covert capabilities than prior Claude models on several of our evaluations, particularly when its reasoning is hidden from the monitor," the system card states. "That said, we believe that low levels of absolute performance on our new covert evaluations support claims that rely on detection of harmful actions in high-stakes situations."

Fable 5.2 has been available to select parties in pre-release for some time, and according to Anthropic, the model has already "solved two open problems in theoretical physics, formalized Fermat's Last Theorem, and supported numerous other research projects."

Also on Monday, AI startup Physical Superintelligence announced a new proof of the Aaronson‑Ambainis conjecture—a pivotal step in establishing theoretical limits for quantum computing. Notably, the lab said the result was achieved by running "a non-public Anthropic model" within PSI's own system. Anthropic neither confirmed nor denied whether Fable 5.2 was the model in question.