Anthropic’s massive 'Claude Mythos' leak sends software names — and crypto — sharply lower
Summary
Anthropic is testing a significantly more capable AI model internally dubbed "Claude Mythos," described as a major step change in performance.
Leaked internal documents, accidentally exposed in a public data store, warned that this new model could drastically increase cybersecurity risks by rapidly finding and exploiting software vulnerabilities, potentially fueling a cyber arms race.
The news caused immediate market reactions: cybersecurity stocks like Palo Alto Networks, Crowdstrike, and Fortinet fell between 4% and 6%, and the iShares Expanded Tech-Software Sector ETF (IGV) dropped nearly 3%, which likely contributed to Bitcoin falling back toward the $66,000 level. The leak also suggested Anthropic is developing a new tier called "Capybara," which would surpass their current most advanced model, Opus.
(Source:CoinDesk)