OpenAI glitch revokes access to AI program for cybersecurity researchers
What's the story
OpenAI has come under fire after a technical glitch led to the revocation of access for some cybersecurity researchers. The issue affected those who were part of the company's Trusted Access for Cyber (TAC) program, which gives vetted users less restricted access to advanced AI models. The issue was reported by multiple researchers on OpenAI's official support forums and X.
Program details
What is the TAC program?
The TAC program is a special initiative by OpenAI to provide trusted researchers with access to the company's most advanced AI models.
These models come with fewer cybersecurity guardrails than those available to regular users.
The idea behind such programs, also offered by companies like Anthropic through their Cyber Verification Program (CVP), is to equip trusted defenders with better tools for reporting bugs and vulnerabilities, with the aim of getting flaws patched faster.
Issue resolution
Access revocation was a temporary glitch, says OpenAI
OpenAI has confirmed that the access revocation was a temporary glitch on its part.
The company said in a tweet, "Due to a technical issue, a limited set of users' access to Daybreak Blue is no longer active, and they will need to re-verify to maintain their access."
It also assured users that steps are being taken to prevent such issues from recurring in the future.
Re-verification process
Affected users have been notified by email
OpenAI has asked the affected researchers to reapply and complete the verification process again.
The company said in its tweet, "Affected users have been notified by email with a path for verification, and our teams are on standby to help."
It also reminded all users that Advanced Account Security would be mandatory by September 1 to maintain Daybreak access.
Tier details
Daybreak Blue offers access to frontier general-purpose models
Daybreak Blue, the latest vetted tier of the TAC program, offers access to "frontier general-purpose models, including GPT-5.6 Sol, with safeguards tailored to authorized defensive security work," according to OpenAI.
The company launched it on August 10 and describes it as the recommended starting point for most defenders.
It supports vulnerability discovery, secure code review, malware analysis, incident response, and patch validation.