Federal authorities arrested Abdulrahman Alfozan, a 35-year-old Saudi national employed as an engineer at OpenAI, on charges of secretly installing a hidden recording device in an executive conference room at the company’s San Francisco office, the U.S. Department of Justice (DOJ) announced on October 13.
The device, concealed in toilet paper and plugged into a USB port behind a television, was discovered by OpenAI employees on September 30, according to court documents. Prosecutors allege Alfozan installed the device around September 28 to intercept oral communications, including discussions about the company’s AI research and proprietary technology. A similar customized device was later found in his personal bag during an investigation.
OpenAI confirmed in a post on X (formerly Twitter) that it had terminated Alfozan and referred the matter to the FBI, which led to his arrest on October 9. The company stated it had no evidence that Alfozan accessed or exfiltrated model weights, source code, or customer data. OpenAI also noted it used its internal AI models to identify the employee responsible and trace the device’s origins. The FBI’s investigation remains ongoing.
Key Details of the Alleged Incident
The recording device, described as sophisticated by prosecutors, contained a microphone and camera capabilities and was encrypted, preventing immediate access to its stored data. Court filings indicate Alfozan allegedly used OpenAI’s own tools, including a productivity tool and chatbot, to research sensitive locations within the office and methods for secretly placing audio equipment. Specific chatbot prompts cited in the complaint include queries about “building, customizing, installing, and audio equipment”.
The executive conference room where the device was hidden was frequently used for highly confidential discussions, including research, product plans, and business development, according to the DOJ. FBI Director Kash Patel stated in a post on X that Alfozan’s alleged goal was to capture information about OpenAI’s AI product development.
Corporate and Government Responses
OpenAI emphasized in its statement that it takes protecting its technology, data, and employees seriously, calling the incident a matter of national importance. The company reiterated that its security program is designed to address threats from insider risks, cybercriminals, and nation-states, and noted that AI can assist in mitigating such risks.
The Saudi embassy in the United States did not immediately respond to requests for comment from multiple outlets. The DOJ’s criminal complaint was made public on October 13, and Alfozan made his initial court appearance in the U.S. District Court for the Northern District of California on the same day. He faces a federal charge of attempting to use a device to intercept oral communications, which carries potential penalties upon conviction.
Broader Context: AI Industry Security Concerns
This case underscores growing security concerns within the AI sector, where companies and governments compete for talent and proprietary technology that could confer economic and strategic advantages. OpenAI’s incident follows heightened scrutiny of its safety protocols, including the recent firing of three safety researchers who alleged a “significant breach of trust” regarding the handling of sensitive information. The company has not provided further details on the researchers’ claims beyond stating that they violated policies.
The arrest also highlights the role of AI in detecting and responding to security threats. OpenAI’s use of its own models to identify the perpetrator and trace the device’s origin is cited as an example of how AI can enhance corporate security measures. However, the incident raises questions about internal vulnerabilities and the potential for insider threats in high-stakes technological environments.