Saudi Engineer Arrested for Planting Eavesdropping Device at OpenAI Headquarters
- Input
- 2026-10-14 11:07:47
- Updated
- 2026-10-14 11:07:47

[Financial News] A high-performance eavesdropping device was found at the San Francisco office of OpenAI, the developer of the ChatGPT artificial intelligence (AI) chatbot, and a Saudi national who worked as an engineer for the company was arrested by U.S. law enforcement authorities.
On the 13th (local time), foreign media outlets including the Associated Press and Agence France-Presse reported, citing a criminal complaint filed with a federal district court in San Francisco, that the FBI and police arrested a 35-year-old engineer on the 9th for allegedly installing an eavesdropping device in a conference room at the San Francisco office.
OpenAI immediately notified the FBI after discovering a sophisticated recording device in a conference room and fired the employee. The arrested suspect appeared in court on the 13th and was charged with attempting to intercept oral communications using electronic and mechanical devices.
According to U.S. law enforcement authorities, the suspect was found to have hidden a USB-shaped recording device wrapped in tissue paper inside the TV in the “executive conference room.”
Although the suspect was not authorized to attend meetings held in that conference room, internal CCTV footage repeatedly captured the suspect loitering nearby. The company grew suspicious and installed an additional camera inside the room; footage clearly captured the suspect returning to check the device they had hidden.
The complaint also cited evidence that the suspect had used internal tools to look up the company building’s address, the conference room’s location and capacity, and meeting schedules in advance. However, it remains unclear exactly what the suspect intended to do with the audio data obtained through eavesdropping.
OpenAI said it had actively used its AI technology to help resolve the incident.
An OpenAI spokesperson told the Associated Press, “We used our own AI model to analyze the device and its source, and as a result identified a specific employee as having installed it a few days earlier.”
The spokesperson added, “So far, we have found no evidence that the employee improperly accessed or leaked outside the company any core information, including model weights, source code, or customer data.”
U.S. federal authorities are now expanding their investigation to determine the detailed motive behind the crime and whether anyone else was behind it.
[email protected] Yoon Jae-jun Reporter