FBI Weighs Economic Espionage Charges Against Ex-OpenAI Engineer Over Saudi Ties in the Conference-Room Recording Device Case

The FBI is investigating whether to pursue economic espionage charges against Abdulrahman Alfozan, the former OpenAI engineer arrested earlier this month for allegedly planting a recording device in a restricted conference room, the Wall Street Journal reported. Investigators are examining whether Alfozan obtained confidential information with the aim of offering it to the Saudi government, even without being recruited to do so, according to a law enforcement official. Trade secret theft charges are also under consideration, though the probe may not produce additional charges. He returns to court on November 6.

The underlying case reads like an insider threat playbook. Alfozan, 35, a Saudi national and U.S. permanent resident who joined OpenAI as an engineer in July 2025, is accused of concealing a small circuit board and microphone, wrapped in toilet paper, behind a wall-mounted television via a USB port in a San Francisco conference room used primarily by OpenAI executives. Prosecutors say it was installed around September 28 and discovered during a countersurveillance sweep on September 30. On October 5, the day he allegedly returned to retrieve it, OpenAI security found a second similar device in a bag he left unattended. FBI agent Lydia Liang identified the device as a Chinese-made Seeed XIAO ESP32-S3 Sense, customized to encrypt whatever it stored.

AI tools themselves feature prominently in the evidence. Prosecutors say Alfozan used a workplace productivity agent to ask about building addresses, meeting rooms, calendars and the locations of "leadership and exec" meetings, and used a chatbot account tied to his spouse to research the scheme, with prompts including how to "build a continuous audio recorder" and "encrypt the SD card."

The new reporting centers on his Saudi connections. Prosecutors say Alfozan exchanged frequent WhatsApp messages in 2025 and 2026 with an unidentified member of the Saudi royal family who sought to recruit him for a Saudi AI initiative, writing "Saudi needs you more then [sic] the US!" Alfozan allegedly replied "100" but said he intended to stay at OpenAI until his equity vested. Their last known exchange came on October 5, the same day as the retrieval attempt, after OpenAI security had already removed the device and notified the FBI.

The flight-risk record weighed heavily in court. Alfozan bought one-way tickets to Saudi Arabia via Istanbul for himself, his wife and their three-month-old son, and was arrested three days before the planned departure. His wife and son later flew to Qatar on business-class tickets purchased by a Saudi military official, whom the defense identified as the wife's brother-in-law and said used no military funds. When the Saudi Consulate in Los Angeles offered to cover Alfozan's entire bail bond, U.S. Magistrate Judge Alex G. Tse asked "Even if I set it at $10 million?" and called the offer a blank check before denying bail.

Alfozan's attorney, William Pernik, said he had received no indication additional charges were being considered and that his client "will fight these charges and allegations in court, not in the press." OpenAI has said it has no indication that model weights, source code or customer data were inappropriately accessed or taken, that the employee was fired, and that it used its own AI models to help analyze the device and identify who planted it. FBI Director Kash Patel has framed the case as part of protecting American innovation, calling it one of the bureau's highest counterintelligence priorities.

Whatever the grand jury decides, the case has already set two precedents. Physical security at AI labs now matters as much as cyber defense, and the logs of internal AI agents and chatbots can become the evidentiary spine of an insider investigation. The labs guarding frontier models are discovering that the threat surface includes the conference room television.