CNBC
Third-party evaluators find vulnerabilities in OpenAI monitoring system during test pilot
Thursday, October 22, 2026
Third-party evaluators identified vulnerabilities in OpenAI's monitoring systems during a test pilot program, according to reporting by TechCrunch. OpenAI CEO Sam Altman announced the company would provide independent evaluators with "employee-like access" to conduct assessments. The evaluators' findings have not been publicly detailed. Altman did not specify a timeline for addressing the identified vulnerabilities or expanding the independent evaluation program.
Original reporting:
