- OpenAI unveiled Daybreak, its latest security project
- It seeks to rival Anthropic Myth in detecting and remediating high-severity vulnerabilities.
- Daybreak will also help businesses build software more securely from the start.
OpenAI has unveiled Daybreak, its answer to the Anthropic Myth and Project Glasswing, sparking a potential cybersecurity arms race between the two companies.
“Daybreak combines the intelligence of OpenAI models, the extensibility of Codex as an agent harness, and our security partners to help make the world safer for everyone,” the announcement said.
The project seeks to work with OpenAI’s industry and government partners by securing software early in the development process, creating a stronger foundation that will, over time, increase the effectiveness of cyber defense.
Daybreak to build software securely from the start
In the blog post, OpenAI summarizes the project in a single sentence: “The goal is simple: accelerate cyber defenders and continuously secure software. »
Daybreak will allow organizations to apply OpenAI’s Codex security to their own repository using an “agent harness”, where it will find, analyze and remediate attack paths and high-impact code.
High-priority vulnerabilities can be analyzed and validated in a secure, isolated environment, “so teams can prioritize real, repeatable issues over noisy alerts.” Codex Security will also enable teams to automate detection and response, increasing efficiency and securing critical vulnerabilities more quickly.
Daybreak therefore seeks to delegate the work of identification and analysis to AI, and to return the results of vulnerabilities, supported by evidence, to human teams. Where Daybreak differs in its approach from Mythos is in building software that is secure from the start and constantly monitoring for vulnerabilities, compared to Mythos’ goal of detecting and mitigating high-severity vulnerabilities at scale.
Daybreak includes three models; GPT-5.5, default for general purpose work; GPT-5.5 with Trusted Access for Cyber, for use in defensive security workflows; and GPT-5.5-Cyber, for specialized workflows including red teaming and penetration testing.
Dane Knecht, CTO at Cloudflare, said: “We’re excited about the potential of OpenAI’s cyber capabilities to bring stronger reasoning and more agentic execution to security workflows. It’s a big step forward for teams to be able to leverage cutting-edge models to not only accelerate speed, but also improve their security posture.”
Follow TechRadar on Google News And add us as your favorite source to get our news, reviews and expert opinions in your feeds.
