By ByteBot
An OpenAI engineer planted a hidden audio device in the company’s executive conference room. OpenAI used its own AI models to identify him. The FBI arrested him twelve days later.
That is the story in three sentences — and it contains more irony than most tech companies manage in a decade.
What Happened
Abdulrahman Alfozan, 35, a Saudi national who joined OpenAI’s AI agents platform team in July 2025, was arrested on October 9 and charged with federal wiretapping. According to a Department of Justice press release, Alfozan allegedly installed a recording device in an executive conference room at OpenAI’s San Francisco office on approximately September 28.
The device — a Seeed Studio XIAO ESP32-S3 Sense, a thumb-sized IoT development board — was hidden inside toilet paper and plugged into a USB port on the back of a conference room television. It had a microphone, a camera sensor, microSD storage, and had been customized with encryption to evade detection. Gizmodo noted the DOJ described it as “sophisticated” — it retails for 3.99.
OpenAI’s technical surveillance team found it on September 30 during a routine sweep. On October 5, surveillance cameras caught Alfozan attempting to inspect the area behind the television despite having no authorization to enter the room. He also left an unattended bag in a printer area that day; when security personnel opened it, they found a second identical device along with his wallet and ID.
OpenAI Used Its Own AI to Catch the Spy
This is the part worth sitting with. OpenAI didn’t just find the device — it used its own AI models to analyze what the device was, trace where it came from, and identify the specific employee who planted it. The company then referred the matter to the FBI.
The same research capability that was being targeted became the tool used to catch the person targeting it.
FBI Director Kash Patel framed the stakes plainly: Alfozan “allegedly concealed a covert recording device inside his employer’s executive conference room to secretly capture sensitive communications about the company’s super intelligence technology.” U.S. Attorney Craig Missakian added: “For us it is a matter of national security.”
What Was at Risk
The targeted conference room wasn’t incidental. According to the DOJ, it regularly hosted discussions involving “confidential technical, commercial, and strategic information, including proprietary research, product plans, and government engagements.” Investigators say Alfozan had used internal productivity tools to research building addresses, meeting room schedules, and where leadership held meetings — reconnaissance that preceded the installation.
OpenAI says it found no evidence that model weights, source code, or customer data were exfiltrated. But the intent — if the charges stand — was to listen in on the most sensitive conversations happening at one of the world’s most closely-watched AI companies.
The Broader Picture
This isn’t happening in isolation. In January, a federal jury convicted former Google engineer Linwei Ding of AI economic espionage — the first U.S. prosecution of its kind — after he stole more than 2,000 files of AI chip infrastructure data. Ding’s method was digital. Alfozan’s was physical. Both targeted AI research. The methods are diversifying.
The same week Alfozan was arrested, OpenAI fired three safety researchers for allegedly sharing confidential information with an external AI evaluation firm. The week also saw more than 400 scientists publish an open letter demanding mandatory AI agent containment standards. October has not been a quiet month for OpenAI.
There is also a question worth raising about physical security. AI companies have invested enormously in securing model weights, hardening APIs, and building red teams. A 4 IoT board tucked behind a television found its way into OpenAI’s most sensitive meeting room for at least two days before discovery. USB ports in conference room hardware have been a documented attack vector for years. The gap between digital security investment and physical security practice at high-value research organizations is real, and this case makes it harder to ignore.
Alfozan faces up to five years in prison if convicted. His detention hearing is scheduled for October 19. TechCrunch first broke the story on October 13.
