OpenAI Engineer Arrested for Allegedly Planting Bug in Executive Conference Room

Federal authorities arrested Abdulrahman Alfozan, an engineer at OpenAI, on Tuesday after accusing him of planting a covert audio recording device in a San Francisco conference room. The criminal complaint, unsealed on Tuesday, alleges that Alfozan concealed the hardware inside a piece of toilet paper and attached it behind a monitor.

Prosecutors state that Alfozan, a 35-year-old Saudi Arabian national and US permanent resident, used an internal OpenAI productivity agent to check the schedule for the targeted room. He also reportedly used a chatbot account linked to his spouse to ask technical questions about building a hidden listening device.

Surveillance Planning and Discovery

Alfozan allegedly spent several weeks planning the operation using artificial intelligence tools before the physical device was installed. According to the court filing, he inserted the bug into a USB port on the back of a television in the executive meeting space.

OpenAI’s security team found the device on September 30 during a routine countersurveillance sweep of the office. Investigators also discovered a second recording device in Alfozan’s bag at the workplace, according to the complaint.

Legal Charges and Employment Status

Alfozan made his initial appearance in US District Court in San Francisco on Tuesday. He faces one charge of using an electronic device to intercept oral communications, which carries a maximum penalty of five years in prison if convicted.

He joined OpenAI in July 2025 and was terminated last Friday, coinciding with his arrest. Alfozan could not be reached for comment regarding the allegations or the court proceedings.

Motive and Data Security

FBI Director Kash Patel stated in a public release that Alfozan aimed to secretly capture sensitive communications regarding the company’s advanced technology. However, neither federal prosecutors nor OpenAI have publicly confirmed a specific motive for the surveillance attempt.

OpenAI released a statement on X asserting that there are no indications the individual accessed or exfiltrated model weights, source code, or customer data. The company emphasized that protecting its technology and people remains a matter of national importance.

What should you do?

Organizations should review access logs for internal AI tools to detect unusual queries about sensitive locations or schedules. Physical security teams should conduct regular countersurveillance sweeps in executive meeting rooms to identify unauthorized hardware.

What to do and how to stay safe: OpenAI

• Monitor internal AI and productivity tool logs for queries that target restricted physical areas or executive schedules.
• Implement routine physical security sweeps of conference rooms to detect unauthorized audio or video recording devices.
• Establish clear protocols for inspecting personal bags and devices in high-security zones to prevent the introduction of surveillance gear.
• Review employee access privileges to ensure that internal scheduling tools are not used to map out sensitive meetings without authorization.

Step-by-step guide: Mobile Device Security: How to Protect Data Beyond Passwords

General security guidance from the Threat Times newsroom. It is not confirmed advice from the organisations named in this story.

Frequently asked questions

Did the OpenAI engineer steal any code or data?

OpenAI stated that there are no indications that the individual accessed or exfiltrated model weights, source code, or customer data.

What charge was the engineer arrested for?

Alfozan faces one charge of using an electronic device to intercept oral communications, which carries up to five years in prison.

How did the employee plan the surveillance?

Prosecutors allege he used an internal productivity agent to check meeting schedules and a chatbot to ask technical questions about building the device.