OpenAI has acknowledged that a technical issue removed Daybreak Blue access for a limited set of users in its Trusted Access for Cyber program, according to The Register. The program is designed for vetted security professionals, and Daybreak Blue is an access tier that gives approved cyber defenders expanded capabilities. The Register reports that participants noticed this week that previously approved status had disappeared without warning. OpenAI's cyber verification page reportedly treated some of them as if they had never been approved, prompting them to begin verification again. The issue also affected tooling access. According to The Register, some users found that Daybreak Blue had disappeared from Codex Desktop and the command-line interface. OpenAI has attributed the removals to a technical issue. In responses cited by The Register, the company said a limited set of users lost Daybreak Blue access and would need to complete verification again. The company said it was working to prevent the issue from recurring. The recovery process appears uneven. The Register says affected users were directed to look for an email titled “Action required: Reverify your Daybreak Blue access,” but some who followed the instructions reported that the verification system would not proceed because their accounts were ineligible. One participant told The Register they had held Daybreak Blue access for more than a month before losing it, then encountered errors saying their identity could not be verified or the account was not eligible. According to that user's account of an exchange with support, OpenAI could not reset the verification state, reinstate the earlier approval, or override the new result. The distinction matters because Trusted Access for Cyber is not a general feature toggle. As The Register describes it, applicants must provide identification and pass verification before receiving access to more capable cyber tooling. The goal is to give legitimate defenders stronger tools for finding and fixing vulnerabilities while limiting access for users who may misuse them. OpenAI did not provide The Register with answers on how many users were affected, what caused the removals, why some previously approved researchers are now being rejected, or whether the company can restore access when reverification fails. The company instead pointed the outlet to a public statement saying a limited number of users were affected by a technical issue and would need to reverify. Who benefits: If OpenAI fixes the process cleanly, vetted defenders regain access to capabilities intended for vulnerability research and defensive work. The program itself benefits if the company can show that verification errors are reversible and auditable. Who's exposed: Researchers relying on Daybreak Blue access are exposed to interruptions while reverification remains unresolved. OpenAI is exposed to trust and support risk if previously approved users cannot get clear remediation.