OpenAI engineer accused of secretly recording executive meetings

A software engineer employed by OpenAI has been arrested and charged by U.S. authorities after investigators alleged he secretly placed a concealed recording device inside an executive conference room at the artificial intelligence company.

The case has drawn attention to the growing security challenges facing leading AI firms as competition intensifies over advanced technologies that many governments and businesses view as strategically important.

According to the U.S. Department of Justice, Abdulrahman Alfozan, a 35-year-old Saudi national and lawful permanent U.S. resident, joined OpenAI in July 2025. Prosecutors allege he installed a hidden surveillance device in a meeting space used by senior company executives at OpenAI's San Francisco offices.

The device was reportedly discovered by company staff at the end of September after it was found connected to a television in the room. Investigators said the equipment had been disguised and concealed from view. Authorities later recovered a second customised device during a search of Alfozan's personal belongings, according to court documents.

OpenAI launches internal investigation

OpenAI said it initiated an internal investigation after the device was discovered and subsequently referred the matter to the FBI.

In a public statement, the company said it used its own artificial intelligence tools as part of the effort to identify the individual allegedly responsible and trace the origins of the equipment.

The company said investigators uncovered evidence suggesting plans to place additional recording devices in other sensitive areas of its offices.

However, OpenAI said its investigation found no evidence that the employee accessed or removed highly sensitive assets such as model weights, source code or customer information.

The employee's contract was terminated following the investigation, the company added.

Federal investigators allege the device was intended to capture confidential conversations involving OpenAI's advanced AI programmes.

FBI Director Kash Patel said the case involved an alleged attempt to record discussions related to the company's most sensitive research activities.

The complaint also alleges that Alfozan used internal AI productivity tools to seek information related to meeting-room schedules and facilities in the weeks before the device was discovered.

Authorities have not publicly disclosed what information, if any, was successfully obtained or transmitted.

Growing security concerns in the AI industry

The incident comes at a time when artificial intelligence companies are facing increasing pressure to protect proprietary research, talent and commercial secrets.

As competition between major AI developers accelerates, concerns have expanded beyond cybersecurity threats to include physical security, insider risks and corporate espionage.

The case is the latest security-related issue to affect OpenAI in recent weeks. The company recently disclosed that it had dismissed several researchers following an internal review of information-handling practices.

If convicted, Alfozan could face a prison sentence of up to five years, as well as financial penalties and supervised release.

Court proceedings are expected to determine whether prosecutors can prove the allegations and establish the intended purpose of the recording devices.

The FBI investigation remains ongoing, and authorities have not announced any additional charges or identified possible accomplices.

Read more:

• OpenAI and Anthropic say they would welcome Australian AI data breach reporting laws
• OpenAI: Rogue agents tried to erase evidence of hacking activity report shows
• OpenAI apologises to Australia over government AI website hack

Tags

News USA OpenAI spyware Saudi Arabia