OpenAI brings in Hacktron to probe its systems
OpenAI has partnered with security firm Hacktron to hunt for vulnerabilities in its systems, focusing on where they connect to third-party services, the attack surface that AI agents keep expanding.
OpenAI announced a partnership with Hacktron, a security firm that builds AI agents for code review, penetration testing and red teaming. Engineers from OpenAI and researchers from Hacktron will pick high-priority areas to test together. The work centres on how OpenAI's systems interact with third-party services.
As part of the deal, OpenAI is expanding Hacktron's access to its frontier models through Daybreak Red.
OpenAI said it first came across Hacktron through its Security Bug Bounty program and was impressed by the impact of the vulnerabilities the firm's researchers found and disclosed. According to OpenAI, Hacktron's published research has uncovered flaws in how applications, cloud infrastructure and identity systems interact. OpenAI argued that this kind of expertise matters more as AI agents link services together in new ways.
The announcement follows two other recent security moves. Earlier the same week, OpenAI said researchers from Cathedral would join its teams for its next security sprint. As we reported previously, the company also more than quadrupled bug bounty payouts for critical vulnerabilities.
OpenAI did not disclose financial terms, how long the partnership will last, which specific systems are in scope for the joint testing, or whether any findings will be made public. For independent researchers, the in-scope systems and the submission process are still listed on the bug bounty program page.