Back to news
marketOpenAI2026-11-06

OpenAI announced on November 6 a partnership with security firm Hacktron to test vulnerabilities in how its systems interact with third parties

OpenAI engineers and Hacktron researchers will jointly pick high-priority areas to test, and OpenAI is granting Hacktron expanded frontier model access through the Daybreak Red program.

On November 6, OpenAI announced a partnership with security company Hacktron to jointly find security vulnerabilities in OpenAI systems. The two sides will focus on how the systems interact with third-party services, and testing of these cross-system connections will be strengthened as AI agents connect to services in new ways.

Under the arrangement, OpenAI engineers and Hacktron security researchers will jointly select high-priority areas to test. OpenAI is also expanding Hacktron's access to frontier models through the Daybreak Red program. The Hacktron team has long built AI agents for code review, penetration testing and red teaming, and previously found vulnerabilities in the interactions between applications, cloud infrastructure and identity systems.

OpenAI said the partnership continues a series of recent security moves, noting that earlier this week it announced researchers from Cathedral would join its next security sprint and that it had raised bug bounty rewards. OpenAI described putting frontier cyber models into the hands of security researchers so they can uncover complex vulnerabilities at the boundaries between interconnected systems.

OpenAIHacktron安全合作Daybreak Red漏洞