Foreign National Caught Bugging OpenAI's Highly Confidential Meeting Room: FBI

SAN FRANCISCO, CA — A now-terminated OpenAI employee is accused of planting a recording device inside the company's executive conference room, according to an FBI investigation announced on Tuesday.

Abdulrahman Alfozan, 35, a Saudi Arabian national, was charged Tuesday with using or endeavoring to use an electronic, mechanical, or other device to intercept oral communications.

It's unclear why Alfozan allegedly planted the recording device inside the room. But the room is frequently used for "highly confidential" discussions and strategic information, including nonpublic information about OpenAI's relationships and government engagements, according to the U.S. Department of Justice.

OpenAI found the bug concealed in toilet paper behind a TV on Sept. 30 and used its model to discover who planted it, according to a statement from the San Francisco-based company and details from the U.S. Department of Justice.

That led the company to Alfozan, who had no authority to be in the room, federal prosecutors said.

Surveillance footage showed Alfozan near the executive conference room multiple times before the device was found, and then checking behind the TV, according to the FBI.

Alfozan also used a productivity tool to make inquiries about OpenAI's meeting rooms, leadership and executive locations and partner meetings, authorities allege.

Federal prosecutors say Alfozan also used his wife's chatbot for prompts about building and installing audio recording devices and live listening.

On Oct. 5, the same day he was caught checking behind the TV, OpenAI security personnel found his unattended bag with his wallet and ID, authorities said. Inside was a similar recording device to the one found behind the TV, authorities allege.

OpenAI officials said they reported Alfozan to the FBI and fired him.

"We referred the matter to the FBI, and we’re grateful to the agents and prosecutors who moved quickly to make an arrest," OpenAI said. "The FBI’s investigation is ongoing and we will continue to support their work."

There's no evidence that Alfozan accessed or exfiltrated model weights, source code, or customer data, according to authorities.

"We take protecting our technology, data, and people seriously, and as a matter of national importance," OpenAI said. "Our security program is designed to protect against a wide range of threat actors, from insider risks to cybercriminals to nation states, and this case is an example of how AI can help address these sophisticated security threats."

Alfozan was arrested on Friday. He now faces a maximum of five years in federal prison, a $250,00 fine and three years of supervised release if convicted, federal prosecutors said.