OpenAI Expands Cybersecurity Efforts With Tools and Open-Source Patching Program

Summary

OpenAI announced Codex Security updates, the full release of GPT-5.5-Cyber, and Patch the Planet, a program to help open-source maintainers validate and patch vulnerabilities with security researchers.

Key points

Updated Codex Security can scan codebases, trace attack paths, validate findings, generate patches, and export results to vulnerability-management pipelines.
Since its March research preview, Codex Security has processed over 30 million commits across more than 30,000 repositories; human reviewers confirmed over 70,000 fixes, with another 500,000 findings resolved automatically.
GPT-5.5-Cyber is available to verified defenders and scored 85.6% on CyberGym, compared with 81.8% for standard GPT-5.5.
Patch the Planet brings researchers equipped with OpenAI tools together with maintainers to validate findings, deduplicate them, and develop patches before submitting them to maintainers.
More than 30 projects have signed on to Patch the Planet, including cURL, Go, Python, Sigstore, and pyca/cryptography.
The Daybreak Cyber Partner Program will let security vendors integrate GPT-5.5 with Trusted Access for Cyber into their products and services.

Article Details

Event Type: OpenAI announcement expanding its cybersecurity initiative with updated tools, a research partnership, and a security vendor partner program.
Impact: OpenAI says the effort prioritizes vulnerability remediation as AI accelerates vulnerability discovery and overwhelms defenders. Since its research preview launched in March, Codex Security has processed more than 30 million commits across over 30,000 repositories; human reviewers confirmed more than 70,000 fixes, and another 500,000 findings were resolved automatically. Access to GPT-5.5-Cyber remains limited to verified defenders.