On April 8, 2026, Anthropic announced the launch of Project Glasswing and introduced its new, ultra-powerful AI model, Claude Mythos Preview. This model significantly surpasses its predecessor, Opus 4.6, in coding and hacking capabilities, possessing the ability to autonomously discover thousands of high-risk zero-day vulnerabilities across all major operating systems and web browsers. The project involves 12 tech giants, including Amazon AWS, Apple, Google, and Microsoft, as well as over 40 key organizations maintaining critical software infrastructure. These entities will gain access to the model for scanning and fixing vulnerabilities in their core systems and open-source software. Anthropic has committed to providing up to $100 million in model usage credits and donating $4 million to the open-source community to support security maintenance. Following the research preview period, the model will be made available for commercial access to participants at a pricing of $25/$125 per million input/output tokens.