Claude Mythos evolved from a leak suggesting Anthropic was testing a substantially more advanced model into a tightly restricted release defined by its exceptional cybersecurity capabilities.
The US government issued an export control directive citing national security concerns to suspend foreign national access to Anthropic's most advanced AI models, Fable 5 and Mythos 5. In response, Anthropic disabled the models for all users, including those in the US.
Anthropic is scaling access to its Mythos AI model to 150 organizations in more than 15 countries, targeting critical infrastructure sectors including power, water, healthcare, and communications where cyberattacks could affect 100 million people. The expansion is part of Project Glasswing, the company's security vulnerability program.
Anthropic released Claude Opus 4.8 on Thursday, upgrading its flagship model with stronger coding and agentic capabilities at existing pricing levels. The company also announced its more powerful Mythos model will roll out to customers in the coming weeks.
OpenAI announced Thursday it is rolling out GPT-5.5-Cyber (codenamed 'Spud') to vetted cybersecurity professionals for finding vulnerabilities and analyzing malware. Security testing shows the model can autonomously execute sophisticated cyberattacks, completing a 32-step corporate network simulation and solving a 12-hour security puzzle in 10 minutes, matching capabilities of Anthropic's Mythos released a month earlier.
The White House has authorized federal agencies to access Anthropic's controversial Mythos AI model for cybersecurity applications, overriding Pentagon security concerns about the model's advanced exploit capabilities.
Venture capitalists are offering to invest in Anthropic at valuations up to $800 billion, following the company's release of the Mythos model and Project Glasswing cybersecurity demonstration. The valuation represents a significant increase from previous rounds.
