OpenAI announced a partnership with Hacktron on November 6, 2026, to find vulnerabilities in its systems, concentrating on connections with third-party services and expanding the security firm’s access to frontier models through Daybreak Red.
OpenAI engineers and Hacktron security researchers will jointly select high-priority areas to test. The work will examine how systems interact as AI agents connect services in new ways, with the lessons used to strengthen OpenAI’s defenses. That gives the partnership a specific target: weaknesses that emerge when applications, identity systems and infrastructure are linked.
A Bug Bounty Relationship Expands
OpenAI became familiar with Hacktron through its Security Bug Bounty program and said it was impressed by the impact of the vulnerabilities the firm’s researchers discovered and disclosed. Hacktron builds AI agents for code review, penetration testing and red teaming, and its research has uncovered vulnerabilities in interactions among applications, cloud infrastructure and identity systems. Those capabilities align directly with the new testing focus.
Hacktron’s September 13, 2026 account provides a concrete example. Its researchers said they chained two critical vulnerabilities on July 25, 2026, to compromise multiple OpenAI employees’ ChatGPT accounts, creating potential access to internal repositories and other connected services. The chain combined remote code execution in the Discourse-hosted community forum’s image-processing path with an OpenAI single sign-on issue, allowing the forum compromise to cross into ChatGPT and Codex and then a connected GitHub integration.
To demonstrate access without reading sensitive internal code, Hacktron used an employee’s Codex account to create a harmless pull request in OpenAI’s internal monorepo, then stopped testing. It reported the issues through OpenAI’s Bugcrowd program and to Discourse. Hacktron’s timeline says OpenAI confirmed a fix roughly 14 hours after the initial report, while Discourse had a fix ready two days after receiving its report.
OpenAI paid Hacktron a $6,500 bounty on September 1, 2026, while clarifying that the award recognized the OpenAI-side finding rather than testing against the explicitly out-of-scope Discourse-hosted forum, according to Hacktron’s reproduced timeline.
Daybreak Red Access and Controls
OpenAI’s August 10, 2026 Daybreak announcement defines Red as access to purpose-trained cybersecurity models for authorized vulnerability research, exploit validation and security testing, and recommends Blue as the starting point for most defenders. Blue provides frontier general-purpose models with safeguards tailored to authorized defensive work. Teams whose authorized work includes advanced vulnerability research, exploit development or red teaming can request Red access.
Access to both tiers is limited to approved individuals and organizations conducting authorized work. The controls include identity verification, account security, monitoring, approved-use restrictions and legal attestations. OpenAI’s documented best practices also call for isolating security workflows from sensitive production systems and the open internet, reviewing agent actions outside the Codex sandbox before execution, adding human oversight for higher-risk workflows, and using scoped permission profiles to enforce authorized boundaries.
Hacktron’s earlier account also says the work was not completely autonomous and that skilled human guidance remained important. The partnership’s defensive value therefore rests on a balance: expanded access to cyber-capable models for deeper testing of interconnected services, matched by enforceable scope, monitoring and human oversight when agents are used.
Credit: Source link
