OpenAI Restricts Access for Cybersecurity Researchers
Access to the Trusted Access Program was revoked due to a technical issue.
1What is the Trusted Access Program?
OpenAI's Trusted Access Program (TAC) provides cybersecurity researchers with limited access to the company's most advanced AI models. The program aims to facilitate the detection and reporting of security vulnerabilities by researchers. TAC offers fewer restrictions compared to regular user access. A similar program, CVP, is offered by Anthropic.
2How Did the Access Revocation Occur?
On August 19, 2026, multiple researchers reported on OpenAI's official support forums and X that their access to the TAC program had been revoked. Researchers encountered a message stating that their identity could not be verified or that their account was ineligible when accessing ChatGPT's Cyber page. OpenAI confirmed that the issue was caused by an error.
3Details of the Technical Issue
In an email to researchers, OpenAI stated that access to Daybreak Blue, a TAC tier, had been revoked due to a technical issue. The company acknowledged that the problem was on their end and negatively impacted the user experience. Researchers were informed that they needed to reapply and complete the verification process.
4Daybreak Blue and Daybreak Red
Daybreak Blue is OpenAI's latest TAC tier, providing access to frontier AI models like GPT-5.6 Sol. This tier is used for vulnerability discovery, secure code review, and incident response. Daybreak Red offers access to models specifically designed for cybersecurity research, allowing authorized security testing.
5Researchers' Reactions
In recent months, both defensive and offensive cybersecurity researchers have complained about restrictions imposed by OpenAI and Anthropic. Researchers argue that these restrictions hinder legitimate work. The access revocation particularly affected researchers outside the U.S. and Europe.
6The Verification Process
To access the TAC program, cybersecurity researchers must submit their IDs and undergo vetting by OpenAI. This process aims to prevent malicious actors from misusing the program. However, technical issues temporarily restricted access for some researchers.
7OpenAI's Statement
In a statement on X, OpenAI announced that access to Daybreak Blue for a limited number of users was no longer active and required re-verification. The company emphasized that this measure was necessary to ensure user security.
8The Role of AI in Cybersecurity
AI models are increasingly playing a critical role in cybersecurity. Initiatives like the Trusted Access Program enable researchers to use these technologies to quickly detect vulnerabilities. However, access restrictions and technical issues can complicate this process.
Original article: cybersecurity | TechCrunch — OpenAI Siber Araştırmacıların Erişimini Kısıtladı