Two researchers at Facebook parent Meta have proposed a new framework approach for dealing with online threats, that uses a shared model for identifying, describing, comparing, and disrupting the ...

Meta's new hyperagent framework breaks the AI "maintenance wall," allowing systems to autonomously rewrite their own logic and scale across tasks without constant human engineering.

Meta makes its Frontier AI Framework available to all The company says it is concerned about AI-induced cybersecurity threats Risk assessments and modeling will categorize AI models as critical, high, ...

Meta, the parent company of Facebook, Instagram, WhatsApp, has suggested in a new policy document that it might not release its internally developed highly capable AI system under certain ...