Daybreak Is OpenAI’s Response To Anthropic’s Claude Mythos

OpenAI has just launched Daybreak, a cybersecurity initiative that’s clearly the company’s competitor to Anthropic’s Project Glasswing. If you’ll recall, Glasswing uses Anthropic’s unreleased AI model, Claude Mythos Preview, to provide its clients’ cyber defense needs. It’s been promising, so far: Mozilla revealed in April that Mythos helped it find and patch 271 vulnerabilities in the latest release of the Firefox browser. OpenAI says Daybreak uses its various AI models, including its specialized security agent Codex. 

In its announcement, the company explained that Daybreak is built around the premise that cyber defense should be built into software from the start and not just revolve around finding and fixing vulnerabilities. Daybreak aims to prioritize high-impact issues and reduce hours of analysis to minutes, to generate and test patches within repositories and to send back results with audit-ready evidence to the clients’ systems. In OpenAI’s example, it asked Codex Security to scan a codebase, validate the highest-risk findings and fix them. 

Find and fix vulnerabilities earlier with Daybreak pic.twitter.com/yobOSWYeWP

— OpenAI (@OpenAI) May 11, 2026

Daybreak will use GPT-5.5 for general purposes and GPT-5.5 with Trusted Access for Cyber for most defensive security workflows, including “secure code review, vulnerability triage, malware analysis, detection engineering and patch validation.” It will also rely on GPT-5.5-Cyber for “preview access for specialized workflows, including authorized red teaming, penetration testing and controlled validation.” OpenAI is already working with several partners under the initiative, including Cloudflare, Cisco, CloudStrike, Palo Alto Networks, Oracle and Akamai.