Anthropic releases Mythos AI disguised as ‘Claude Fable 5' for everyone

Anthropic has officially released Claude Fable 5 for everyone.

This marks the public availability of the version of Anthropic’s most powerful Mythos-class AI model. Previously, companies labelled the model as “too dangerous” for general use.

In April, the tool was released privately, which caused alarm among technology, finance, and government leaders.

Acknowledging the risks, Anthropic said: “Releasing a model this capable comes with risks. Without safeguards, Fable 5’s capabilities in areas like cybersecurity could be misused to cause serious damage. We’ve therefore launched the model with safeguards that mean queries on some topics will instead receive a response from our next-most-capable model, Claude Opus 4.8.”

Limited access for cyber defenders

Anthropic also launched Claude Mythos 5 for a small group of cyberdefenders and infrastructure providers, the same model without cybersecurity restrictions. The tool is deployed via Project Glasswing in association with the US government.

In its early testing, Mythos identified more than 10,000 critical security flaws.

Pricing and Availability

The model is priced at $10 per million input tokens and $50 per million output tokens. This is priced less than half the cost of the previous version.

The private valuation of the Anthropic is reportedly near $1 trillion as it prepares for an unexpected public offering.