Anthropic Expands AI Cybersecurity Program to Include Healthcare Organizations
Anthropic is expanding Project Glasswing — its AI-powered cybersecurity initiative — to 150 additional organizations, adding healthcare to the program for the first time. The program uses Claude Mythos Preview, a frontier model withheld from public release due to dual-use security concerns, to scan partner codebases for vulnerabilities. The initial cohort of roughly 50 organizations identified more than 10,000 high- or critical-severity flaws. The new round spans 15-plus countries and brings in healthcare, power, communications, and hardware sectors — many participants are vendors and open-source maintainers whose software governments and large institutions depend on. Anthropic estimates that a major attack on most Glasswing partners could affect more than 100 million people.
Healthcare systems depend on layered vendor software stacks with little visibility into upstream security posture. Glasswing’s expansion signals that AI-powered code scanning is crossing into enterprise healthcare — operators should evaluate whether their critical vendors are auditing their own code with equivalent rigor.
While we aim to share useful and relevant resources, we do not guarantee the accuracy of content on this site or any external links. Views and opinions expressed in referenced content do not necessarily reflect those of Healthcare Growth Strategies.
What is Anthropic's Project Glasswing?
Project Glasswing is Anthropic's program that uses Claude Mythos Preview — a frontier AI model not publicly released — to scan software codebases for security vulnerabilities. The program's initial 50 partners identified more than 10,000 high- or critical-severity flaws. Healthcare is now included in the expanded second cohort of 150 organizations.
Why is Anthropic expanding Project Glasswing to healthcare?
Healthcare was underrepresented in the first round of Glasswing partners. The expansion reflects the sector's growing dependence on third-party software and the outsized impact a cyberattack could have — Anthropic estimates attacks on most partners could affect more than 100 million people.
How does Claude Mythos Preview differ from publicly available Claude models?
Claude Mythos Preview is a frontier AI model that Anthropic has intentionally withheld from public release due to its dual-use cybersecurity capabilities. It is deployed exclusively through Project Glasswing, where it scans partner codebases under a controlled, vetted program — not available through the standard API or consumer products.