New AI Model Exposed Security Flaws in Major Operating Systems and Web Browsers

Anthropic is launching a new AI-powered cybersecurity solution called Project Glasswing, which aims to identify vulnerabilities in large-scale systems with minimal human intervention. Partnered with leading tech companies such as Nvidia, Google, Amazon Web Services, Apple, Microsoft, and others, the model is touted as a game-changer for organizations seeking to bolster their security posture. The breakthrough AI model, codenamed Claude Mythos Preview, was initially developed in-house but is now being made available exclusively to launch partners due to concerns over its potential misuse. According to Anthropic’s cyber lead, Newton Cheng, the AI can detect and flag critical vulnerabilities with unprecedented accuracy, enabling companies to prioritize remediation efforts and stay ahead of emerging threats. By leveraging machine learning algorithms and vast amounts of data, Project Glasswing is poised to revolutionize the cybersecurity landscape, empowering organizations to proactively manage risk and protect sensitive information from sophisticated attacks.