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

The Verge / 4/8/2026

📰 NewsSignals & Early TrendsIndustry & Market MovesModels & Research

共有:

Key Points

Anthropic is launching Project Glasswing, a cybersecurity initiative with major partners including Nvidia, Google, AWS, Apple, and Microsoft to help large organizations flag software vulnerabilities with minimal human effort.
The effort is built on access to Claude Mythos Preview, a general-purpose model Anthropic is not planning to publicly release because of security concerns.
Anthropic says its frontier red team identified security problems across every major operating system and web browser, motivating the model’s use in vulnerability discovery.
The project is positioned to support both private enterprise and potentially government security workflows, shifting AI use toward automated vulnerability detection and reporting.
The announcement highlights how model deployment and release decisions are being directly influenced by cybersecurity risk assessments rather than purely by capability benchmarks.

Photo collage showing mouse cursors circling like sharks in a tablet screen.

Anthropic is debuting a new AI model as part of a cybersecurity partnership with Nvidia, Google, Amazon Web Services, Apple, Microsoft, and other companies. Project Glasswing, as it's called, is billed as a way for large companies, and potentially even the government, to flag vulnerabilities in their systems with virtually no human intervention.

Anthropic is offering its launch partners access to Claude Mythos Preview, a new general-purpose model that it's not currently planning to publicly release due to security concerns. Newton Cheng, the cyber lead for Anthropic's frontier red team, told The Verge that the model will ideally give cyber …

Read the full story at The Verge.