Several security researchers report that OpenAI unexpectedly revoked their access to a restricted program designed to ease certain limitations on using its AI tools for cybersecurity work. OpenAI has acknowledged the situation, attributing it to an error. On Wednesday, multiple researchers took to OpenAI’s official support forums and X to share that their access to the Trusted Access for Cyber (TAC) program had been cut off. According to these individuals, when they visited ChatGPT’s Cyber page, they encountered a notification stating that their identity could not be verified or that their account "is ineligible at this time."

TAC is a specialized initiative through which OpenAI provides vetted researchers with access to its most advanced AI models, featuring fewer cybersecurity safeguards than those available to everyday users. Anthropic runs a comparable program known as the Cyber Verification Program (CVP). The core purpose of both TAC and CVP is to equip trusted defenders with superior tools, enabling them to identify and report bugs and vulnerabilities to companies, thereby accelerating the patching process. At the same time, these programs aim to keep cybercriminals and malicious actors from accessing such models, preventing them from discovering flaws and crafting exploits to breach corporate systems. To gain entry into TAC, cybersecurity researchers must submit identification and undergo OpenAI’s vetting process.

At this stage, the exact reason for these revocations remains unclear, as does the number of affected individuals. One researcher noted receiving an email from OpenAI stating that their access to Daybreak Blue—the latest vetted tier of TAC—had been revoked "due to a technical issue affecting a limited number of users." In a separate thread on OpenAI’s forums, a researcher mentioned that after contacting official support, the company similarly cited a "recent technical issue" that had caused some users to lose access to Daybreak Blue. In both instances, OpenAI advised the researchers to reapply and go through the verification process again. Reports suggest the affected individuals are based in the U.S. and Europe, hinting that the revocations may be confined to specific regions.

Daybreak Blue, introduced by OpenAI on August 10, is the newest tier for individual researchers, granting access to "frontier general-purpose models, including GPT-5.6 Sol, with safeguards tailored to authorized defensive security work." According to OpenAI, it "is the recommended starting point for most defenders, supporting vulnerability discovery, secure code review, malware analysis, incident response, and patch validation."

Alongside this, the company also launched a higher tier called Daybreak Red, which offers access to models built specifically for cybersecurity research. This tier allows vetted users to conduct "authorized vulnerability research, exploit validation, and security testing."

In recent months, both defensive and offensive security researchers have voiced frustration over the guardrails imposed by Anthropic and OpenAI, arguing that these restrictions hinder their ability to carry out legitimate work. OpenAI has since provided additional comment on the matter.