NewsProduct
Anthropic's Fable 5.2 and Mythos 5.2: about 60 percent fewer cyber blocks, stronger bug finding, and a gated route for defenders
Anthropic released Claude Fable 5.2 and Mythos 5.2 on November 9, 2026. They are the same underlying model; Fable carries cybersecurity and biology safeguards and Mythos does not. On Fable 5.2, classifiers block high-risk cyber work such as penetration testing, exploit generation and binary-based vulnerability work, and a blocked request is answered by Opus 4.8. Anthropic says the safeguards are as robust as Fable 5.1's with fewer unnecessary blocks: about 60 percent fewer cybersecurity blocks in early testing. Security professionals can apply for Mythos 5.2 through the Cyber Verification Program, and Claude Enterprise customers can use it through Claude Security. On Anthropic's own CVE Rediscovery test of 1,829 published vulnerabilities, Mythos 5.2 rediscovered 89.6 percent against 81.3 percent for Opus 5.5 with safeguards off; Anthropic says one security firm found that one in six of the bugs it confirmed had been missed by every other frontier model it tested, and Mozilla's Brian Grinstead told Help Net Security that Mythos 5.2 roughly doubled the vulnerability discovery metrics of Mythos 5.1 in Firefox scanning. In a deliberately weakened sandbox, Anthropic says Mythos 5.2 used credentials or files placed outside it in 3.1 percent of more than 1,800 scenarios, against 21 percent for Mythos 5.1 and 39 percent for Mythos 5. Nearly all the figures are Anthropic's own and have not been independently confirmed.
Why it matters
The models that find bugs best are gated by who you are, and the gate is moving: fewer blocks for ordinary security work, a verification route for defenders. Expect more AI-found vulnerabilities and more patches, and treat vendor benchmark figures as claims until outside testers repeat them.