Key takeaways
- OpenAI engineer Abdulrahman Alfozan, 35, was arrested Oct. 9 and charged with trying to intercept oral communications, according to the Justice Department.
- Prosecutors say a recorder was hidden behind a TV in an executive conference room and a second device was found in his bag on Oct. 5.
- OpenAI says it has no indication he accessed or took model weights, source code or customer data, and a detention hearing is set for Oct. 19.
An OpenAI engineer has been charged with planting a hidden recording device in an executive conference room at the company's San Francisco offices. The OpenAI engineer arrested in the case, Abdulrahman Alfozan, 35, is a Saudi national who had worked at OpenAI since July 2025, according to the U.S. Justice Department. He was arrested on Friday, Oct. 9, and made his first appearance in federal court on Tuesday, Oct. 13.
Federal officials are treating the case as more than a workplace dispute. "For us it is a matter of national security," U.S. Attorney Craig Missakian said in the Justice Department's announcement. The case lands at a moment when AI labs are guarding model research and business plans as closely as defense contractors guard their designs.
What happened
According to the criminal complaint, as summarized by the Justice Department and the San Francisco Standard, OpenAI staff found an audio recording device on Sept. 30. It was plugged into a USB port on the back of a television in an executive conference room and hidden inside toilet paper. The company estimates it had been installed around Sept. 28. Its contents were encrypted.
The Justice Department says that room was often used for highly confidential discussions, including research, product plans and business development. Prosecutors say Alfozan was not authorized to enter the room or attend meetings there.
Then, on Oct. 5, things moved quickly. Prosecutors allege Alfozan entered the conference room and appeared to look behind the television. The same day, he allegedly left a bag unattended in a different OpenAI building. Security staff found his wallet, his ID and a second recording device similar to the first one inside, according to CBS News.
The digital trail
Some of the most striking details in the complaint involve AI tools themselves. Prosecutors allege that before the device was installed, Alfozan used a workplace productivity tool to ask about OpenAI building addresses, meeting room capacities, calendars and where "leadership and exec" meetings took place, CBS News reports.
The complaint also points to a chatbot account linked to his spouse that shared identifiers with his own account. Between Sept. 24 and Sept. 30, that account allegedly received prompts about building, installing, testing and remotely accessing a continuous audio recorder. The San Francisco Standard reports that on Sept. 27 and 28, the account allegedly asked about the specific device that was later found, including its MAC address.
These are allegations. Alfozan has not been convicted of anything.
The charge and possible penalties
Alfozan faces one count of using or attempting to use a device to intercept oral communications. If convicted, he faces up to five years in prison, a $250,000 fine and three years of supervised release, according to the Justice Department.
He remains in custody. A detention hearing is set for Oct. 19 before U.S. Magistrate Judge Alex G. Tse, the San Francisco Standard reports. The FBI is investigating, and the case is being handled by federal prosecutors in the Northern District of California with help from the Justice Department's National Security Division.
His defense attorney, William Pernik, told CBS News that his client is presumed innocent. He urged people to avoid "jumping to conclusions at this very early stage of the case when very little information is known" and warned against trying to "politicize" the case based on nationality.
What OpenAI says
OpenAI says its technical surveillance countermeasures team found the device during a routine sweep of the conference room. Employees then used the company's own AI models to figure out what the device was and where it came from before handing the matter to the FBI, the San Francisco Standard reports.
The company described the recorder as sophisticated and built to avoid detection, with strong encryption and a way of moving data meant to slip past monitoring. OpenAI also says it has evidence that Alfozan intended to place more recording devices in other sensitive spaces.
On the question most customers will care about, OpenAI told CBS News that "we have no indications of this individual inappropriately accessing or exfiltrating model weights." The company also said it has seen no sign he took source code or customer data, according to the San Francisco Standard.
OpenAI says it coordinated the timing of his firing with the FBI. Starting Sept. 30, it gradually restricted his access, then suspended his physical and network access, placed him on leave and secured his work equipment. He was fired on the day of his arrest. The company called the protection of its technology a matter "of national importance."
Why this case matters
AI labs now hold some of the most valuable secrets in the tech industry: training methods, unreleased products, partner deals and the weights of models that cost billions of dollars to build. That makes them targets for competitors, criminals and foreign governments.
FBI Director Kash Patel said protecting U.S. AI innovation is one of the bureau's highest counterintelligence priorities. Placing the case with national security prosecutors signals that the government sees AI company secrets as a strategic asset, not just corporate property.
The case also fits a wider pattern of insider risk at tech companies. Just this week, the Justice Department said North Korean IT workers used laptop farms to get hired at about 70 US companies. Different scheme, same lesson: the threat is often someone who already has a badge or a login.
OpenAI has had a turbulent stretch on the people side as well, including fired safety researchers publishing an open letter disputing claims against them.
Who is affected
OpenAI and its partners
Executives and partners who met in that room will want to know what, if anything, was captured. OpenAI says the recording was encrypted and found within about two days of installation, but it has not said publicly what the device recorded.
Other AI companies
Every major lab will be reviewing its physical security after this. Expect more bug sweeps, tighter room access and closer monitoring of internal tool usage. The fact that OpenAI's own AI helped identify the device may push other firms to use similar tools for security work.
Employees in tech
Workers at AI labs may see stricter rules about which rooms they can enter, what they can bring into the office and how their activity in internal tools is logged.
What companies can learn from it
You do not need to run a frontier lab to take something from this case. A few practical steps apply to most businesses:
- Sweep sensitive rooms. Regular checks for unknown devices in boardrooms and executive offices are cheap compared with a leak.
- Watch the ports. USB ports on TVs, conference phones and displays are easy hiding spots. Disable or cover the ones you do not need.
- Limit room access. Badge controls that match a person's role make it easier to spot someone where they should not be.
- Review internal tool logs. Unusual searches for executive calendars or meeting locations can be an early warning sign.
- Lock down accounts. Strong account security is the basic layer under all of this. Our guide to protecting your online accounts in 10 minutes is a good place to start.
What to watch next
The next date to watch is the Oct. 19 detention hearing, which will decide whether Alfozan stays in custody. Prosecutors may also file an indictment, and the FBI says its investigation is ongoing. Key open questions include whether anyone else was involved, who, if anyone, was meant to receive the recordings, and whether investigators recover any audio from the devices.
For more on security and big tech, visit our tech news section.
Frequently asked questions
Who is the OpenAI engineer arrested by the FBI?
He is Abdulrahman Alfozan, 35, a Saudi national who had worked at OpenAI as an engineer since July 2025, according to the Justice Department.
What is he accused of?
Prosecutors say he hid an audio recording device behind a TV in an executive conference room and had a second device in his bag. He is charged with trying to intercept oral communications.
Did he steal OpenAI's model weights or customer data?
OpenAI says it has no indication that he improperly accessed or took model weights, source code or customer data.
What penalty could he face?
If convicted, he faces up to five years in prison, a $250,000 fine and three years of supervised release.
Share this story
Spotted an error? Tell us
