A new Anthropic model found security problems ‘in every major operating system and web browser’

Anthropic is debuting a caller AI exemplary arsenic portion of a cybersecurity concern with Nvidia, Google, Amazon Web Services, Apple, Microsoft, and different companies. Project Glasswing, arsenic it's called, is billed arsenic a mode for ample companies, and perchance adjacent the government, to emblem vulnerabilities successful their systems with virtually nary quality intervention.

Anthropic is offering its motorboat partners entree to Claude Mythos Preview, a caller general-purpose exemplary that it's not presently readying to publically merchandise owed to information concerns. Newton Cheng, the cyber pb for Anthropic's frontier reddish team, told The Verge that the exemplary volition ideally springiness cyber …

Read the afloat communicative astatine The Verge.