OpenAI has just released its answer to Claude Mythos

OpenAI is starting Ripean AI approach that focuses on detecting and identifying vulnerabilities before they occur. Daybreak uses the Codex Security AI agent was launched in March create a threat based on the organization’s rules and look for potential attack paths, determine potential vulnerabilities, and identify those at greatest risk.

Its launch comes barely a month after its partner Anthropic announced Claude Mythosa type of security-oriented AI that claims to be too dangerous to be released publicly and is only shared privately as part of its operations, known as Project Glasswing. However, that did not stop several illegal parties from getting access.

However, OpenAI does not currently have a similar security feature. Like Glasswing, Daybreak isn’t built on just one AI model – OpenAI he says “Daybreak brings brands OpenAI, Codex, and our security partners.”

Daybreak also includes special examples of cyber, including GPT-5.5 with Cyber ​​security feature ​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​ and GPT- or GPT-5.5-Cyberwhich was first released last week. OpenAI also says it is working with “industry and government partners” as it plans to “deploy more advanced cyber applications.”