Several security researchers report that OpenAI suddenly cut off their access to a limited program designed for cybersecurity work, and the company says the issue was a mistake on its end.

On Wednesday, multiple researchers took to OpenAI's official support forums and X to say their access to the Trusted Access for Cyber (TAC) program had been revoked. When they opened the ChatGPT Cyber page, they saw messages saying their identity could not be verified or that their account was "ineligible at this time."

TAC gives vetted researchers access to OpenAI's most advanced AI models with fewer cybersecurity guardrails than standard users get. A similar program, the Cyber Verification Program (CVP), is run by Anthropic. The goal of both is to let trusted defenders use powerful models to find and report bugs faster, while keeping malicious hackers from abusing the tools.

To join TAC, researchers must submit an ID and pass OpenAI's vetting process. It's still unclear why access was pulled or how many people were affected. TechCrunch spoke to five researchers who experienced the problem. One said OpenAI emailed them that access to Daybreak Blue—the latest vetted tier of TAC—was revoked "due to a technical issue affecting a limited number of users." The email added, "This was an issue on our end, and not the user experience we want to deliver."

In an OpenAI forum thread, a researcher said support also cited a "recent technical issue" and asked them to reapply and complete verification. All five researchers TechCrunch interviewed live outside the U.S. and Europe, hinting the revocations may be region-specific.

OpenAI referred TechCrunch to a tweet stating that a "limited set of users' access to Daybreak Blue is no longer active and they will need to re-verify to maintain their access."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, launched August 10, grants access to frontier models like GPT-5.6 Sol with safeguards for defensive security work—covering vulnerability discovery, code review, malware analysis, incident response, and patch validation. OpenAI also introduced a higher tier, Daybreak Red, for authorized vulnerability research and exploit validation.

Recently, both defensive and offensive security researchers have complained that guardrails from OpenAI and Anthropic block legitimate work. OpenAI later updated its response.

Source: TechCrunch