OpenAI has confirmed that a technical fault ejected some vetted security researchers from its Trusted Access for Cyber (TAC) programme, and its recovery process is now rejecting attempts by some of them to regain the access they had already been approved for. Participants found their previously verified status had disappeared, with the verification page inviting them to start the process afresh as though they had never been cleared, while the Daybreak Blue tier offering expanded capabilities vanished from Codex Desktop and the command-line interface.

OpenAI told affected users, via a community forum post and a statement on X, that "a limited set of users" had lost access due to a technical issue and would need to reverify, directing them to an email titled "Action required: Reverify your Daybreak Blue access." However, several researchers reported that reverification failed, with the system declaring their accounts ineligible or unable to confirm their identity; one had held access for over a month before losing it. OpenAI support reportedly told at least one user it could neither reset the verification state nor restore or override their prior approval, and the company did not answer The Register's questions on how many users were affected, the cause, or whether access could be restored.