Unconfirmed: An X account claimed on 12 November that three OpenAI executives were fooled on a 17-minute call by Russian pranksters posing as Ukraine's foreign intelligence chief. Automatica Press has not confirmed the details of this incident.
@ns123abc wrote on 12 November that three of OpenAI’s top Pentagon defense and policy executives were deceived in a call with individuals pretending to be Ukraine's foreign intelligence chief Source @ns123abc on X1. The poster asserts that the executives believed they had met the caller previously and were cooperative, unaware that the person on the other end had their camera off for the entire duration.
The post further alleges that the 17-minute recorded conversation included discussions regarding OpenAI and Pentagon military contracts, offensive AI capabilities, Trump’s foreign policy, and Elon Musk. According to the source, OpenAI confirmed it was pranked. No statement from OpenAI or law enforcement has been found to verify these claims.
This claim arrives amid a broader landscape of unverified social media reports concerning OpenAI. Earlier this week, Automatica reported on separate allegations that OpenAI had released eight GitHub projects for building AI agents, which also lacked independent verification Source Automatica Press (earlier coverage)2.
The specific allegation of a social engineering breach involving high-level defense contractors raises significant security concerns. However, without official confirmation from OpenAI or the Department of Defense, the authenticity of the recording and the identity of the perpetrators remain unestablished. The source provides no evidence beyond the textual description of the event.
An OpenAI statement clarifying the nature of the call, or the publication of the full recording, would confirm or kill the rumour.