OpenAI launches Daybreak cybersecurity initiative to rival Anthropic’s Project Glasswing

OpenAI has unveiled a new cybersecurity initiative called Daybreak, marking the company’s latest move into AI-powered digital defence services. The programme is widely seen as OpenAI’s response to Anthropic’s Project Glasswing, which has attracted attention for using advanced artificial intelligence to identify and repair software vulnerabilities.

The launch highlights the growing competition among major AI firms to position their models as essential tools for cyber defence. As organisations face increasingly complex online threats, technology companies are racing to offer automated systems capable of identifying weaknesses, analysing risks and speeding up software protection processes.

OpenAI focuses on proactive cyber defence

OpenAI said Daybreak is designed around the idea that cybersecurity should be integrated into software development from the beginning rather than treated as a separate process after systems are deployed. According to the company, the initiative aims to reduce the time needed for security analysis from hours to minutes while helping organisations prioritise the most serious threats.

The company explained that Daybreak can generate and test patches directly within software repositories, then return results with audit-ready evidence. OpenAI said this process is intended to improve efficiency for security teams and reduce the burden of manual investigation.

As part of its announcement, OpenAI demonstrated how its Codex Security agent could scan a codebase, identify high-risk vulnerabilities, validate the findings and automatically generate fixes. The company suggested this workflow could help developers address threats more quickly while maintaining oversight of the repair process.

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

— OpenAI (@OpenAI) May 11, 2026

Daybreak uses several of OpenAI’s AI models to support different security tasks. GPT-5.5 will handle general operations, while GPT-5.5 with Trusted Access for Cyber is designed for defensive workflows such as secure code review, malware analysis, vulnerability triage, detection engineering and patch validation.

OpenAI also revealed that it will provide access to GPT-5.5-Cyber for specialist security activities. The company said this model is intended for authorised red teaming, penetration testing and controlled validation exercises, allowing organisations to test the resilience of their systems under supervised conditions.

Competition grows in the AI cybersecurity sector

The launch of Daybreak comes as AI companies continue expanding into the cybersecurity market, an area viewed as one of the most commercially significant uses of generative AI technology. Anthropic has already gained attention with Project Glasswing, which uses its unreleased Claude Mythos Preview model for cyber defence applications.

Interest in Anthropic’s work increased after Mozilla disclosed in April that Mythos had helped identify and patch 271 vulnerabilities in the latest version of the Firefox browser. The results suggested that advanced AI systems could significantly accelerate vulnerability detection and repair compared with traditional manual approaches.

OpenAI’s announcement appears to position Daybreak as a direct alternative for enterprises seeking AI-assisted security tools. By introducing specialised security-focused models and automated patch-generation features, the company is seeking to establish itself as a major player in the rapidly evolving cyber defence market.

Industry analysts have increasingly noted that cybersecurity has become a strategic battleground for AI developers. Businesses are under pressure to strengthen digital security as cyberattacks become more sophisticated, while security teams often struggle with staffing shortages and growing workloads.

AI-powered systems are being promoted as a way to reduce response times and improve threat detection. However, concerns remain over the reliability and oversight of automated tools, particularly when they are granted access to sensitive systems or tasked with generating software fixes independently.

OpenAI has not disclosed pricing details or broader availability for Daybreak, but the company indicated that it is already collaborating with several major technology and cybersecurity firms through the initiative.

Major technology companies join Daybreak programme

OpenAI said several industry partners are already involved with Daybreak, including Cloudflare, Cisco, CrowdStrike, Palo Alto Networks, Oracle and Akamai. The partnerships suggest that the initiative is being positioned for large-scale enterprise and infrastructure protection rather than consumer-focused security products.

The inclusion of established cybersecurity firms may also help OpenAI address concerns about trust and reliability in AI-driven defence systems. By working alongside recognised security providers, the company appears to be aiming for broader acceptance among enterprise customers.

Several of the companies participating in Daybreak already provide security infrastructure used by governments, financial institutions and global businesses. Their involvement indicates that OpenAI is targeting organisations that manage high-value systems and large-scale online operations.

The launch also reflects a broader shift within the technology industry, where AI is increasingly being integrated into security operations centres and software development pipelines. Automation tools are now being used not only for detecting threats but also for analysing malicious code, monitoring network activity and recommending remediation steps.

While AI-driven cybersecurity tools continue to improve, experts have warned that attackers are also adopting similar technologies to develop more advanced threats. This has created what many analysts describe as an escalating technological arms race between cybercriminals and defence providers.

OpenAI’s entry into the market through Daybreak adds further momentum to that competition. As companies continue investing in AI-based security systems, the effectiveness and reliability of these tools are likely to come under increasing scrutiny from both customers and regulators.