Federal prosecutors have charged Abdulrahman Alfozan, a Saudi national who joined OpenAI as an engineer in July 2025, with allegedly attempting to intercept private conversations at the company. The indictment alleges that Alfozan used artificial intelligence tools to research office layouts and develop recording equipment, a development that highlights how the same technologies companies are racing to release can be repurposed for criminal planning.
The case emerged after OpenAI discovered a "surreptitious audio recording device" plugged into a USB port in the back of a television within an executive conference room around September 30. In a post on X, OpenAI stated that it utilized its own AI models to help understand the device and trace its origin. According to an FBI affidavit, investigators allege that Alfozan used a "productivity agent" to query basic information about OpenAI's offices. Specific prompts cited in the documents include requests to identify when a meeting was recorded, list meeting rooms in the company that are six feet or larger, and list the day's meetings for a specific room.
The complaint further details that an AI chatbot account tied to Alfozan's spouse was used to instruct a model to build a "continuous audio recorder." The prompts also requested adding "browser access an controls" and encrypting the memory card where recordings were stored. The legal filing did not specify whether any sensitive conversations were successfully captured by the device.
This incident is part of a broader pattern in which AI tools have assisted individuals in carrying out alleged crimes. Prosecutors noted that they identified a Nebraska man's use of ChatGPT prior to a bank robbery, which included asking the model for assistance with stripping and repairing a handgun; that individual was sentenced to more than 10 years in prison in August. Additionally, authorities reported last year that a man used ChatGPT to help plot the explosion of a Tesla Cybertruck outside the Trump International Hotel in Las Vegas.
Alfozan's arrest follows other high-profile cases involving tech employees and trade secrets. In 2024, the Justice Department indicted Linwei Ding, a former Google software engineer, for stealing AI technology secrets and sharing them with two Chinese companies. A federal jury found Ding guilty on several counts of economic espionage and theft of trade secrets earlier this year.
The case underscores a shifting threat landscape where AI models create security challenges through containment breaches as well as by enabling less sophisticated actors to build spying devices. While these tools may make espionage easier to execute, they also provide investigators with new digital trails to retrace steps, adding complexity to both corporate security and law enforcement efforts.