OpenAI Says Russian Pranksters Posing as Ukraine Spy Chief Fooled Its National Security Team
Russian pranksters posing as Ukraine's intelligence chief got three OpenAI national security officials talking military AI on a 17-minute call, The Intercept reports. OpenAI says it investigated weeks ago, confirmed the prank and took steps to stop a repeat.
OpenAI has spent the past year building a national security team stacked with former government officials. According to The Intercept, three of them just spent 17 minutes talking military AI with a man they believed was Ukraine's foreign intelligence chief. He was a Russian prankster with his camera off.
The Intercept reported on Wednesday night that the Russian comedy duo known as Vovan and Lexus posed as Rustem Umerov, head of Ukraine's Foreign Intelligence Service, and landed a video call with Sasha Baker, who runs national security policy at OpenAI, Caroline Zier, the company's lead for strategic initiatives, and Ben Crampton, who leads its European national security work. The pranksters uploaded a recording to Rumble. OpenAI has confirmed the call was a prank.
The company says it knew weeks ago.
"Weeks ago, we launched a thorough investigation into the call's authenticity. We acted internally and with the relevant authorities to conclude that it was a prank conducted by individuals with a history of targeting officials and other companies. We took action to prevent similar situations from happening in the future." Nate Evans, OpenAI spokesperson, to The Intercept
OpenAI has not said what that action was, or when the call took place. The Ukrainian Embassy in Washington did not immediately respond to The Intercept.
This was not a junior staffer clicking a bad link. The Intercept notes that Baker and Zier held senior Defense Department posts and Crampton worked at the European Union and the United Nations. According to the report, Baker and Zier both said on the call that they had met Umerov before and were pleased to speak with him again. The fake Umerov appears to have kept his camera off the entire time.
Offense, defense and the wrong man
According to The Intercept, the conversation covered OpenAI's military support to Ukraine in its defense against Russia, including the possibility of providing offensive capabilities. Asked about selling offensive military AI services, Crampton did not say yes. He said OpenAI wants "as deep a relationship as we possibly can with Ukraine," but may not get there straight away.
"There's a general reluctance on the part of many people who are senior in the company to get too involved in offensive uses of our technology in a military context. Part of our job is to explain that offense and defense are not necessarily distinguishable from each other under certain circumstances." Ben Crampton, OpenAI, on the recorded call
That second sentence is the one worth sitting with. A policy staffer described part of his job as persuading colleagues that defense and offense can blur, to someone he thought was a foreign spy chief. Nothing in the report shows OpenAI has supplied Ukraine with offensive tools, and Crampton's own words suggest it has not.
The officials did describe other contacts, according to The Intercept. Baker said the company had been in touch with the Pentagon about aiding Ukraine. They mentioned work with the Ukrainian government around Daybreak Blue, which The Intercept describes as an OpenAI model for fixing computer vulnerabilities, and what Crampton called early contact with Ukraine's Ministry of Defense, mainly through a Washington-based colleague. Those are remarks on a call they thought was real, not a published program.
The setup was almost lazy, by the pranksters' account. Vladimir Kuznetsov, the Vovan half of the duo, told The Intercept they first contacted a senior Ukrainian official in San Francisco while posing as Umerov. That official, Kuznetsov said, was already trying to build ties with OpenAI and offered to arrange the meeting.
Vovan and Lexus have pranked public figures since 2011, from Elton John to John McCain. The Intercept says their stunts have often lined up with Russian foreign policy priorities and that they have been accused of working with the Russian government, which they deny.
OpenAI is courting defense work openly. It holds a $200 million contract with the U.S. Department of War, The Intercept reports, and it recently hired former Army Secretary Dan Driscoll. A team chasing that kind of work should know who is on the other end of the call. This time, two comedians with a dark webcam got there first.

