OpenAI engineer arrested over hidden spy device in the office

The U.S. Department of Justice has announced the arrest of an OpenAI engineer accused of planting a hidden recording device in an executive conference room at the company's San Francisco office.

What happened

Abdulrahman Alfozan, 35, who had worked at OpenAI since July 2025, was detained on October 9 and charged with illegally intercepting oral communications. According to the DOJ, OpenAI's security team found the recorder on September 30, plugged into a USB port on the back of a television in an executive conference room. The company determined the device had been concealed in toilet paper, installed around September 28, and held encrypted data.

That room was regularly used for meetings about OpenAI's proprietary research, product plans, technical capabilities, business relationships, and government engagements.

The evidence

Security cameras allegedly caught Alfozan near the conference room on multiple occasions, even though he was not cleared to attend meetings held there. Using a workplace productivity tool, he reportedly searched for details about the company's offices, meeting rooms, executive locations, and schedules.

Most telling are prompts to a chatbot account tied to his spouse, which asked how to build and conceal recording devices — how to make a continuous audio recorder, modify Wi-Fi connectivity, improve live listening, and encrypt an SD card. Between September 27 and 28, that same account submitted prompts directly referencing the discovered device, including its unique MAC address.

On October 5, cameras again recorded Alfozan in the room, looking behind the television. The same day, security staff found a second, similar recorder in a bag he had left unattended in another room.

AI turned against its own engineer

A curious twist: OpenAI leaned on its own models to investigate. "We used our models to understand the device and where it had come from, identifying that a specific employee had planted it a few days earlier," the company said. It fired the employee and referred the case to the FBI.

OpenAI stresses it has found no signs of improper access to or exfiltration of model weights, source code, or customer data. If convicted, Alfozan faces up to five years in prison and a $250,000 fine.

Source: Bleeping Computer