Anthropic expands access to AI model for cybersecurity

Key facts

• Anthropic is expanding access to its AI model, Mythos.
• Mythos is designed to discover cybersecurity vulnerabilities.
• Access is being granted to 150 additional organizations globally.
• The expansion covers more than 15 countries.
• Anthropic previously deemed Mythos too dangerous for public release.

Anthropic is expanding access to its AI model, Mythos, to 150 additional organizations globally across more than 15 countries. Mythos is specifically designed to identify cybersecurity vulnerabilities. The company had previously restricted access to this model, considering it too dangerous for widespread public release. This move suggests a potential shift in Anthropic's approach to sharing powerful AI tools, possibly balancing risk with the benefits of broader security research.