OpenAI Cites Technical Error After Vetted Security Researchers Lose Access to Cyber Tools
Multiple cybersecurity researchers recently discovered that their access to OpenAI’s Trusted Access for Cyber (TAC) program had been revoked without prior warning. OpenAI has since confirmed that the problem was caused by an internal technical error affecting a limited group of users, advising those impacted to reapply for verification.
What Happened
Researchers began reporting on OpenAI’s official support forums and on X that they could no longer access the platform’s specialized cybersecurity features. When visiting ChatGPT’s Cyber interface, users encountered prompts stating their identity could not be verified or that their account was currently ineligible.
According to messages shared by affected researchers, OpenAI acknowledged the disruption via email and support replies, describing it as an internal technical issue affecting a limited number of users rather than an intentional policy enforcement. The company directed impacted users to undergo the verification process again to restore access to Daybreak Blue, a recently launched vetted tier within the program.
While the exact number of affected accounts remains undisclosed, researchers who spoke with TechCrunch indicated they are based outside the United States and Europe, indicating the glitch may have impacted international accounts.
Key Highlights
- System disruption: A technical issue on OpenAI’s side deactivated access to the Daybreak Blue tier for an unspecified number of cybersecurity defenders.
- Verification requirement: OpenAI confirmed that affected individuals must resubmit their details and complete re-verification to regain access.
- Program purpose: The Trusted Access for Cyber program provides vetted experts with advanced frontier models featuring fewer cybersecurity guardrails, enabling vulnerability research, malware analysis, and patch validation.
- Regional observations: Impacted researchers interviewed about the issue noted that they were located outside of Europe and the United States.
Why This Matters
Programs like OpenAI’s TAC and Anthropic’s Cyber Verification Program (CVP) exist to assist authorized security defenders in identifying software vulnerabilities and assisting organizations with patch validation before malicious actors can exploit flaws. Advanced AI models typically feature strict safeguards to prevent the generation of malicious code; vetted programs balance these protections by offering authorized researchers tailored access. When system errors unexpectedly lock legitimate researchers out, it temporarily stalls authorized defensive analysis and vulnerability reporting workflows.
What to Watch Next
OpenAI has directed affected participants to re-complete their identity verification. Observers will be monitoring how quickly OpenAI processes these reapplications and whether any further technical adjustments are made to safeguard researcher accounts from unexpected status revocations.
Frequently Asked Questions
What is OpenAI’s Trusted Access for Cyber (TAC) program?
TAC is a specialized initiative that provides vetted security professionals with access to advanced AI models equipped with modified guardrails tailored for legitimate defensive security tasks, such as finding bugs, analyzing malware, and validating software patches.
What is Daybreak Blue?
Introduced by OpenAI on August 10, Daybreak Blue is a vetted access tier designed for individual defensive researchers. It provides access to frontier models, including GPT-5.6 Sol, with safeguards optimized for authorized defensive operations.
Why was researcher access revoked?
OpenAI stated that the loss of access was not intentional, but rather the result of an internal technical error affecting a limited set of users.
Source: TechCrunch and OpenAI Community Forum reports.