A Saudi OpenAI engineer was arrested for planting a hidden recording device in an executive conference room to capture superintelligence talks, federal officials say.

CNBC reported that the Justice Department announced the arrest of Abdulrahman Alfozan, a 35-year-old national of Saudi Arabia and OpenAI employee, in connection with installing a sophisticated covert audio recording device inside an executive conference room at the company’s San Francisco office.

Alfozan was arrested on Oct. 9. OpenAI terminated him. The FBI framed the case as an effort to protect American advances in superintelligence technology from secret monitoring.

Federal officials say Alfozan was not authorized to enter the room where the device was found. OpenAI said the employee planted it a few days before discovery, and the bureau moved quickly on the arrest.

Device hidden in plain sight inside restricted room

The New York Post reported that OpenAI found a surreptitious audio recorder stuffed into a USB port on the back of a television in the conference room and wrapped in toilet paper. The device was described as Seeed Studio branded, with encrypted contents.

Justice Department materials described footage placing Alfozan near the restricted conference room on multiple days before the find. On or about Oct. 5, he went into the room. A bag belonging to him turned up in the OpenAI building around the same time. Inside were his wallet, his ID, and another recording device similar to the one in the executive space.

OpenAI stated it found no signs he had improperly accessed or removed model weights, source code, or customer data. The company said it treats protection of its technology, data, and people as a matter of national importance.

"At this time, we have no indications of this individual inappropriately accessing or exfiltrating model weights, source code, or customer data."

That assurance came as the company stressed the stakes of guarding frontier AI work inside U.S. labs.

Chatbot prompts tracked planning steps

Fox News reported that federal prosecutors say Alfozan used a ChatGPT account tied to his wife to research how to build a continuous audio recorder, enable live listening, and manage downloading or deleting recordings. His wife has not been charged.

The Justice Department described a chatbot account associated with Alfozan’s spouse that shared multiple identifiers and attributes with his own account. Prompts covered building, customizing, installing, audio recording, testing, obfuscating, and remote accessing a device, plus continuous audio recording and Wi-Fi use.

Supporting accounts place the planting window near late September, shortly after assembly instructions appeared in the account history. Core DOJ materials center the entry and bag recovery on or about Oct. 5, with the arrest four days later.

FBI casts case as counterintelligence priority

FBI Director Kash Patel tied the alleged conduct directly to sensitive company discussions. In a statement carried across coverage of the arrest, Patel said 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."

Just the News reported Patel’s further warning that protecting American superintelligence innovation is essential to the nation’s economic strength and national security, and that the FBI views that mission as one of its highest counterintelligence priorities.

Special Agent in Charge Scott Schelble of the FBI’s San Francisco Field Office struck the same note, saying any attempt to illicitly obtain protected information from U.S. superintelligence and technology companies threatens both American innovation and the national security interests those advances support.

The bureau’s posture here fits a wider pattern of aggressive case-building. Readers following FBI investigative work on high-stakes criminal cases will recognize the same focus on locking down intent and method before an arrest.

Alfozan faces up to five years in prison and a $250,000 fine if convicted, according to charging details carried in multiple accounts of the case.

Saudi ties and venture role draw fresh scrutiny

Alfozan had worked as an OpenAI engineer since July 2025. Raed Ventures, a Riyadh-based firm focused on the Middle East and North Africa, listed him as a partner and venture partner. The firm announced H.H. Prince Khalid Bin Bader Al Saud as a venture partner on Feb. 6, 2025, and named Alfozan two days later. A Raed announcement cited prior roles for Alfozan at Meta, Microsoft, and SAP, and studies in computer science and engineering at MIT. Raed did not immediately respond to a request for comment.

The Washington Examiner reported that the arrest, revealed by OpenAI and the Justice Department, raised immediate questions about U.S. efforts to keep AI leadership at home while competition from Saudi-linked capital and talent grows.

Foreign nationals holding sensitive roles inside frontier AI labs create obvious exposure. When an employee lacks clearance for an executive room yet appears on camera near it for days, then leaves a second matching device with his ID, the failure is operational as much as personal.

That enforcement picture also tracks with other recent bureau actions, including moves that expanded the FBI’s fraud most-wanted list against people accused of stealing from taxpayers and investors.

OpenAI has faced separate internal turbulence. Last week the company defended firing three safety researchers, alleging a significant breach of trust and violations of clear policies on handling sensitive information. Those researchers, in a letter to leadership, said communications around their exits made former colleagues afraid to speak. That dispute sits beside the arrest rather than inside it. The recording-device case stands on its own facts: unauthorized access, a planted bug, encrypted audio gear, and chatbot research into continuous listening.

Protecting the lab floor is national security work

Superintelligence research now sits at the center of economic and military competition. A hidden microphone in an executive conference room is not a prank. It is an attempt to vacuum up strategy, technical capability, and any government relationship talk that happens behind closed doors.

Federal investigators treated it that way. OpenAI called the FBI. Agents moved. The employee was cut loose. The public statements from Patel and the San Francisco field office put the mission in plain language: American AI innovation is a national asset, and stealing a look at it is a counterintelligence problem.

Scrutiny of how the bureau runs sensitive probes remains part of the public record, from DOJ oversight of FBI assessments to older case files. Here the priority is clearer. A foreign national on the payroll allegedly wired a room he had no right to enter.

Companies racing to build the next generation of AI cannot treat badge access and conference-room hygiene as afterthoughts. Neither can the agencies charged with keeping that work from walking out the door in a toilet-paper wrap.

When a lab holds technology the FBI calls vital to economic strength and national security, a single unauthorized device is one device too many. Accountability starts with who got in, what they planted, and how fast the system caught it.