KARACHI – OpenAI has partnered with cybersecurity research firm Hacktron to identify security vulnerabilities in its AI systems, with particular attention to connections involving third-party services.
The company announced the OpenAI Hacktron partnership on November 6, 2026, as part of efforts to strengthen security while AI tools become more powerful and connect with external applications.
Under the agreement, Hacktron researchers will work directly with OpenAI engineers to identify potential weaknesses and prioritize security testing.
OpenAI will also expand Hacktron’s access to its advanced AI models through the Daybreak Red programme, which supports authorized cybersecurity research.
Why OpenAI Chose Hacktron
The partnership follows earlier security research in which Hacktron identified serious vulnerabilities affecting OpenAI’s systems.
In July 2026, Hacktron researchers discovered weaknesses that allowed them to compromise several OpenAI employee accounts during authorized security testing.
They also demonstrated access to an internal OpenAI code repository before reporting their findings.
According to Hacktron’s research disclosure, OpenAI confirmed a fix approximately 14 hours after receiving the initial report.
OpenAI said Hacktron’s previous work through its bug bounty programme influenced its decision to deepen the collaboration.
Hacktron develops AI-powered tools for code reviews, penetration testing and security assessments.
Read More : ChatGPT helped wife plan murder.
OpenAI Expands AI Security Testing
The announcement follows several other cybersecurity measures introduced by OpenAI.
On November 2, the company announced that security researchers from Cathedral would participate in its next internal security testing programme.
OpenAI also increased rewards for critical vulnerabilities under its Security Bug Bounty programme by more than four times.
These initiatives aim to identify weaknesses before malicious actors exploit them.
Hacktron CEO Zayne Zhang said combining human expertise with advanced AI capabilities could help researchers uncover complex vulnerabilities across connected systems.
Read More: Why are WhatsApp messages stuck on Zong?
What the Partnership Means for ChatGPT Security
As AI assistants connect with external applications, cloud platforms and business systems, security becomes increasingly important.
A vulnerability involving one connected service could potentially expose other systems or sensitive information, depending on the permissions involved.
The OpenAI Hacktron partnership aims to investigate these risks and strengthen protections across interconnected services.
However, the companies have not announced a timeline or disclosed how many vulnerabilities the new collaboration has identified or resolved.
For now, the partnership represents another step in OpenAI’s effort to improve cybersecurity as AI systems become more widely used.
Read More: National CERT warns of 10 Node.js security vulnerabilities.
Key Points:
- OpenAI announced its Hacktron cybersecurity partnership on November 6, 2026.
- Researchers will focus on vulnerabilities involving third-party services.
- Hacktron previously discovered serious security weaknesses affecting OpenAI employee accounts.
- OpenAI will expand Hacktron’s access to advanced AI models through Daybreak Red.
- The company has also increased critical bug bounty rewards by more than four times.
