OpenAI Engineer Accused of Planting Listening Device in San Francisco Meeting Room
Al Jazeera Staff
A Saudi citizen working as an engineer at OpenAI has been arrested by the FBI and charged with planting a hidden recording device in an executive meeting room at the company's San Francisco headquarters. The incident underscores growing security concerns around sensitive AI research as competition for talent and technology intensifies globally.
A Saudi citizen working as an engineer at OpenAI has been arrested by U.S. federal authorities and charged with planting a hidden recording device in a meeting room at the company's California headquarters, according to the U.S. Department of Justice.
According to the Federal Bureau of Investigation (FBI), Abdulrahman Alfozan, 35, has worked at OpenAI since July 2025. On September 30, the company discovered a device plugged into a USB port behind a television in an executive meeting room at a building in San Francisco.
After the discovery, OpenAI placed a camera in the room. According to the FBI affidavit, the camera recorded Alfozan entering the room last week and checking behind the television — where the listening device had been placed.
The incident comes as companies and governments are competing fiercely for talent and AI technology capable of reshaping regional economies and strategic positions, heightening security concerns around sensitive AI research.
Who is Alfozan and what is he charged with?
According to records, Alfozan has been a lawful permanent resident of the United States since September 2020. A LinkedIn page bearing his name states that he studied at the Massachusetts Institute of Technology (MIT) and the University of California, Berkeley, and previously worked at Meta and Microsoft. Al Jazeera said it could not independently verify the information on the LinkedIn page.
Before joining OpenAI, Alfozan was an investment partner at Riyadh-based Raed Ventures — a fund that invests in seed-stage and early-stage startups in Saudi Arabia and the Middle East and North Africa region.
On October 13, Alfozan made his first court appearance on a charge of "using or attempting to use an electronic, mechanical, or other device to intercept oral communications." It is not yet clear whether he has a defense attorney. He is scheduled to return to court on October 19. If convicted, Alfozan could face up to 5 years in prison, a fine of up to $250,000, and 3 years of supervised release after completing his prison term.
What happened at OpenAI?
According to the FBI, the hidden device was the size of a fingertip, equipped with a microphone, and capable of connecting to a camera. The device was wrapped in toilet paper and placed behind a television in the meeting room. According to the investigation, the device was planted around September 28, and the stored data had been encrypted and was inaccessible. Investigators later found a similar device in Alfozan's personal bag.
In a post on X, OpenAI said there was evidence the employee intended to place additional listening devices in other sensitive areas of the office to secretly record meetings. The company stated there is currently no indication that this individual accessed or stole model weights, source code, or customer data without authorization, and said it has terminated the employee's contract.
The FBI affidavit also alleges that Alfozan used an AI productivity tool to repeatedly ask about building addresses, meeting rooms, meeting room capacities, and the company's meeting schedules, despite not having access to those areas or participating in meetings there.
U.S. Attorney General Todd Blanche said in a statement on X that the case "is being treated as extremely serious" and that the theft of intellectual property related to advanced superintelligence technology is a serious threat to national security.
Has OpenAI previously faced allegations related to espionage?
In a post on X explaining how its systems were used to detect Alfozan, OpenAI said its security program is designed to counter a range of threat actors, from insider risks to cybercriminals and nation-states, and that this case is an example of how AI can help address sophisticated security threats.
However, AI technology is not only used to detect espionage but can also be used to conduct it. In February 2025, an investigation by the AP news agency revealed that major U.S. tech companies facilitated Israel's tracking and killing of people in Gaza and Lebanon through AI and computing services. The AP reported that the Israeli military purchased advanced AI tools from tech companies, including OpenAI, through Microsoft's Azure cloud platform, to support surveillance. OpenAI denied having a relationship with the Israeli military.
OpenAI has also faced criticism for signing an agreement with the U.S. Department of Defense in February of this year to provide technology for U.S. military use in classified environments, including intelligence gathering. OpenAI's rival Anthropic refused to sign a similar agreement unless the Pentagon banned mass surveillance of U.S. citizens.
There have been several incidents in which OpenAI's models were said to have "gone out of control," escaping their testing environments and infiltrating other organizations and even government websites, fueling calls for tighter regulation of unchecked AI surveillance.
