Anthropic Launches Claude Fable 5 With Safety Caps

What did Anthropic release?

Anthropic launched Claude Fable 5, a Mythos-class model now available for general use. According to Anthropic's announcement, Fable 5 outperforms every model the company has previously made publicly available. It shows strong results across software engineering, knowledge work, vision, and scientific research.

Anthropic also notes a clear pattern: the longer and more complex a task, the wider Fable 5's lead over its other models.

What safeguards are in place?

Fable 5 is capable enough in areas like cybersecurity that misuse is a real concern. Because of this, Anthropic launched it with safeguards that redirect certain queries to Claude Opus 4.8 — its next-most-capable model — instead of Fable 5.

You might also like

Supermicro Raises $7B to Fund $39B in AI Server Orders

Three Mile Island Restarts as Crane Clean Energy Center

DOE Opens RFI for AI Data Centers on Federal Land

Urenco USA Plans Multi-Billion Uranium Expansion

Anthropic tuned these safeguards conservatively. That means they will sometimes block harmless requests. On average, the safeguards trigger in fewer than 5% of sessions. The company says it will work to cut false positives as more capable models arrive in the coming months.

If you want a closer look at Claude Opus 4.8 itself, see the Claude Opus 4.8 review covering whether it's an incremental upgrade or hype.

Who gets access to Mythos 5?

Alongside the general release, Anthropic is launching Claude Mythos 5 for a small group of cyberdefenders and infrastructure providers. As Anthropic details, Mythos 5 uses the same underlying model as Fable 5. The key difference: safeguards are lifted in some areas.

Mythos 5 will first be deployed through Project Glasswing. That project runs in collaboration with the US government and serves as an upgrade to Claude Mythos Preview.

A separate report from sources.news noted that Anthropic was preparing to release a public version of Mythos ahead of the official announcement.

Frequently asked questions

Anthropic's confirmed next step: improve the safeguards and reduce false positives as more capable models arrive in the coming months.