OpenAI partners with Hacktron to hunt vulnerabilities across connected services

What did OpenAI and Hacktron announce?

OpenAI announced on November 6 that it is partnering with Hacktron, a security firm that builds AI agents for code review, penetration testing and red teaming, to find vulnerabilities in OpenAI's systems. The two teams will choose high-priority areas to test together, with a focus on how OpenAI's systems interact with third-party services.

Hacktron confirmed the announcement in its own blog post, saying its researchers will work with OpenAI's engineers to prioritize testing of connections with third-party services, with broader frontier-model access through Daybreak Red. "The future of cybersecurity won't be defined by who has the largest security team, but by who can best combine human creativity with AI-driven capabilities," said Zayne Zhang, CEO of Hacktron.

The work gives the partnership a specific target: weaknesses that emerge when applications, identity systems and infrastructure are linked, the kind of boundary problems that become more common as AI agents connect services in new ways.

How did the relationship start?

OpenAI met 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, according to coverage of the announcement. Hacktron's research has uncovered vulnerabilities in interactions among applications, cloud infrastructure and identity systems, capabilities that align directly with the new testing focus.

Hacktron's account of its earlier work provides a concrete example of the risk profile. 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 image-processing path of the Discourse-hosted community forum 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.

What is Daybreak Red access?

Daybreak Red is part of OpenAI's cyber defense initiative for authorized security work. 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, while 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, with controls including identity verification, account security, monitoring, approved-use restrictions and legal attestations.

The Hacktron partnership expands the firm's access to those cyber-capable models, deepening the relationship announced this week. 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.

Why does this matter for AI security?

The partnership is a signal about how frontier labs expect security work to happen: not just with human researchers, but with cyber-capable AI agents testing the seams between services. OpenAI has been expanding this area, from the Daybreak initiative to its GPT-6 cyber model and its pipeline that found hundreds of Linux kernel exploits.

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.

The July account compromise chain is the cautionary detail worth remembering. It shows that the surface OpenAI is now asking Hacktron to test — connections between the community forum, single sign-on, ChatGPT, Codex and GitHub — is exactly where real damage has already been demonstrated.

What should readers take away?

OpenAI is formalizing a relationship with a security firm that has already found and reported serious bugs in its connected systems, and it is giving that firm more access to its frontier cyber models under the Daybreak Red umbrella. The focus is explicitly on third-party connections, which are the parts of the stack where the July compromise chain lived.

For anyone building on OpenAI's platform, the practical takeaway is that the boundary between OpenAI services and external integrations is an acknowledged risk area, and OpenAI is investing in testing it. The bug bounty relationship that produced the July findings is now an ongoing partnership with expanded model access, which changes the threat actor calculus for anyone trying to find similar issues first.