Exclusive
Mystery of Anthropic's Unreleased Mythos AI Model Cracked

Main Street
570 followers
474 Video Views·Apr 17, 2026
Anthropic, one of Silicon Valley's leading AI firms, has built a new model called Claude
Mythos but refused to publish it, saying it is too dangerous to be released to the public.
Instead, they are only giving access to the Claude Mythos preview
to a handful of big companies to help them find and fix security vulnerabilities.
In a report published on Thursday, Bloomberg detailed what Nicholas Carlini,
a well-known AI researcher, has found out about the unreleased AI model.
Carlini discovered Mythos was able to autonomously create powerful break-in tools including against
Linux, the open-source code that underpins most of modern computing, the report says.
What Mythos does is similar to a bank robbery, getting past security protocols and through the
front door of networks and breaking into digital vaults that give it access to online treasures.
And the model seemed to grow stronger each day,
discovering system flaws that only the world's best hackers can find,
posing the biggest cybersecurity risk.
Within hours of getting the model, we knew it was different,
Logan Graham, who runs Anthropic's Frontier Red team, told Bloomberg.
