OpenAI said it found a “sophisticated recording device” hidden in one of its offices and used its own AI to identify who planted the device.
“We used our models to understand the device and where it had come from, identifying that a specific employee had planted it a few days earlier,” OpenAI wrote in an X post on Tuesday. “We referred the matter to the FBI, and we’re grateful to the agents and prosecutors who moved quickly to make an arrest. The FBI’s investigation is ongoing and we will continue to support their work.”
A statement from the Justice Department said that the employee in question is allegedly 35-year-old Abdulrahman Alfozan, a Saudi national. Alfozan was arrested on October 9 and charged with using an electronic device to intercept oral communications. Alfozan made his first appearance in federal court on Tuesday, the statement said.
A complaint filed in the Northern District of California said that an employee found a USB device surreptitiously placed on the back of a television located in an “executive conference room.” The device was concealed in toilet paper, the complaint said.
Photos attached to the court documents and DOJ statement showed the device positioned behind what appears to be a television. The complaint said the device consists of a “thumb-sized development board with a detachable camera sensor, digital microphone, and microSD storage.”
Department of Justice
The complaint further said that CCTV footage showed Alfozan near the conference room on multiple dates prior to the device’s discovery.
“According to the complaint, meetings in the executive conference room where Recording Device 1 was discovered frequently involved discussions of highly confidential technical, commercial, and strategic information, including proprietary research, product plans, and business-development opportunities, as well as nonpublic information about OpenAI’s technical progress and strategic decisions, capabilities, priorities, government engagements, and commercial relationships,” the DOJ said.
Department of Justice
Prosecutors allege that Alfozan used a productivity tool to inquire about OpenAI’s building addresses and meeting rooms, and used a chatbot account associated with Alfozan’s spouse to submit prompts related to building and hiding an audio recording device.
Some of the queries to the productivity agent included a list of “today’s meetings in this room” and “what meeting was this presented at and when was it recorded,” the complaint said.
Department of Justice
A LinkedIn account with the same name as the defendant in the complaint showed that Alfozan was an engineering lead at OpenAI and is a partner at Raed Ventures, a Riyadh-based venture capital firm that invests in early-stage tech startups across the Middle East and North Africa.
A contact associated with the LinkedIn account and a spokesperson for Raed did not immediately return a request for comment. Alfozan’s attorney did not return a request for comment.
OpenAI said that it found no evidence that the employee improperly accessed or manipulated model weights, source code, or customer data. The employee was fired, OpenAI said in the social media post.
OpenAI told Business Insider that its models were used to help with the technical investigation of the incident, including analyzing forensic images of the device’s storage. It also said it has evidence that the employee intended to plant more recording devices throughout OpenAI’s offices.
OpenAI said in the X post that the case was an example of how AI can help address “sophisticated security threats.”
The announcement comes on the heels of a cluster of firings at OpenAI unrelated to the recording device. In the prior case, the company let go of three safety and alignment employees who it alleged had “violated clear policies on handling sensitive information.”
The workers, Jasmine Wang, Mikita Balesni, and Tomek Korbak, published a letter disputing some of the claims circulating about their conduct. They wrote that they were concerned about the lack of clarity around their firings, as well as the way they might chill transparency at the company.
OpenAI stood by the dismissals. The company published an internal note that said, “Our internal investigation uncovered a significant breach of trust beyond what’s outlined in the letter they published and we stand by the decision to not continue their employment.”
