OpenAI launched Codex Security in research preview as a system designed to help identify and validate software vulnerabilities, and Google Chrome release notes have since directly credited the project for multiple browser flaws. Public reporting says the credits were limited to entries that explicitly named Codex Security, indicating the tool has moved from preview into externally visible vulnerability discovery work.
Chrome release notes attributed three high-severity issues to Codex Security in 2026: CVE-2026-15903, an out-of-bounds read/write in V8; CVE-2026-76045, a WebGL use-after-free; and CVE-2026-17658, a V8 use-after-free discovered by Duc Nguyen of Calif.io in collaboration with OpenAI Codex Security. The credited findings span two of Chrome's most security-sensitive components, suggesting the project contributed to the discovery of memory-safety bugs that could have exposed users to browser compromise.

See affected versions and whether adversaries are exploiting it.
4 events from the most recent confirmed update back to the earliest known activity.
Chrome's August 2026 releases credited Duc Nguyen of Calif.io in collaboration with OpenAI Codex Security for CVE-2026-17658, a high-severity use-after-free vulnerability in V8.
Chrome's July 2026 releases directly credited OpenAI Codex Security for CVE-2026-76045, a high-severity use-after-free vulnerability in WebGL.
OpenAI announced Codex Security as being available in research preview. This is the earliest explicitly dated event in the provided references.
Chrome release notes for a July 16 stable update directly credited OpenAI Codex Security for CVE-2026-15903, a high-severity out-of-bounds read/write vulnerability in V8.
Vulnerabilities, threat actors, malware, products, organizations, and breaches Mallory has linked to this story.
See whether adversaries are exploiting this yet, and where the affected versions run in your environment.
2 references tracked. Mallory keeps watching after this page renders.
Map indicators from this story to your assets and identify affected systems in minutes.
Every observed campaign, victim, and pivot linked to actors named in this story.
Malware, exploits, and IOCs connected to the activity described here.
YARA, Sigma, and Snort rules deployed to your SIEM as soon as they’re published.
Get matching new stories delivered to your team as they break — not the next morning.
Ask questions about this story and take action on the answers.