Researchers say OpenAI revoked their access to limited cyber program

Several security researchers have reported that OpenAI suddenly revoked their access to the Trusted Access for Cyber (TAC) program, a limited-access platform that provides more freedom in using AI tools for cybersecurity research. The issue was confirmed by OpenAI, which attributed it to an error. Multiple researchers on OpenAI's support forums and social media platform X reported receiving messages stating that their identity could not be verified or that their account was ineligible at the time.

TAC offers vetted researchers access to OpenAI's most advanced AI models with fewer cybersecurity restrictions compared to regular user access. The goal is to enable trusted defenders to report bugs and vulnerabilities to companies, prompting faster patching, while preventing cybercriminals from exploiting these models. The program requires researchers to submit identification and undergo vetting by OpenAI.

It remains unclear why these researchers' access to TAC was revoked and how many people have experienced the issue. Five researchers contacted by TechCrunch reported facing this problem. One researcher shared an email from OpenAI stating that their access to the Daybreak Blue tier, the latest vetted version, was revoked "due to a technical issue affecting a limited number of users."

OpenAI cited a "recent technical issue" that caused some users to lose access to Daybreak Blue, asking the researchers to reapply and complete the verification process.

All contacted researchers are non-U.S. and non-European residents, suggesting the revocations may be limited to specific regions. OpenAI referred TechCrunch to a tweet confirming that a "limited set of users' access to Daybreak Blue is no longer active and they will need to re-verify to maintain their access." Daybreak Blue provides access to frontier general-purpose models, such as GPT-5.6 Sol, with tailored safeguards for defensive security work, including vulnerability discovery, secure code review, malware analysis, incident response, and patch validation.

The company also introduced a higher-tier program, Daybreak Red, granting access to models designed specifically for cybersecurity research, enabling authorized vulnerability research, exploit validation, and security testing.