Who was the OpenAI engineer who allegedly bugged a company conference room?

The OpenAI logo is seen displayed on a cell phone in front of an image on a computer screen generated by ChatGPT’s Dall-E text-to-image model, Friday, Dec. 8, 2023, in Boston. (Michael Dwyer / Associated Press) By Suhauna Hussain Staff Writer Follow Oct. 17, 2026 6 AM PT 4 7 min Click here to listen to this article Share via Close extra sharing options Email Facebook X LinkedIn Threads Reddit WhatsApp Copy Link URL Copied! Print 0:00 0:00 1x This is read by an automated voice. Please report any issues or inconsistencies here . See more from the L.A. Times in Google Search. Set us as preferred Federal prosecutors allege an OpenAI engineer used AI chatbots to help build and conceal a recording device in a San Francisco conference room used for executive and government meetings. The company says its security team used AI to identify the culprit, who now faces a federal charge. The case has heightened Silicon Valley and Washington concerns about insider threats and foreign access to sensitive technology as the Trump administration casts AI leadership as a national security priority. Build a continuous audio recorder. Develop speech-triggered recording. Add Wi-Fi connectivity. Encrypt memory card. These are some of the prompts an engineer arrested by federal authorities last week is accused of submitting to an artificial intelligence chatbot to program a recording device hidden in an OpenAI conference room. In a criminal complaint filed in U.S. District Court for the Northern District of California and unsealed Tuesday, federal prosecutors allege the engineer then stashed the recording device in an executive conference room in one of the company’s San Francisco offices used for leadership meetings and briefings with government officials. Advertisement The filing describes how OpenAI’s security team on Sept. 30 found the thumb-size recording device folded in a length of toilet paper and plugged in to the back of a television set in the conference room. The company said they then used AI to identify the alleged culprit. Details are emerging about the accomplished engineer at the center of the accusations, Abdulrahman Alfozan, a Saudi national who studied at MIT and long worked in Silicon Valley, and who has spoken about hoping to eventually return to work in the tech industry back home. The Justice Department arrested him Oct. 9 and charged him with illegally using an electronic device to intercept oral communications. Notes of foreign subterfuge in the case, denied by Alfozan’s attorney, have captured the attention of Silicon Valley companies and been seized upon as a major security concern by the Trump administration as it pushes for technological dominance in the AI boom. The administration, meanwhile, has rebuffed calls to regulate AI despite warnings of catastrophe by tech executives themselves, saying it would hamper economic growth and competition. Advertisement “This serious situation underscores why winning the global super intelligence race is a national security prerogative,” said Jay Clayton, director of national intelligence, in a news release this week. He said his agency is “committed to safeguarding America’s cutting-edge intellectual property from foreign threats.” Business OpenAI bot’s rogue attack rattles industry leaders, policymakers and consumers What will happen to the world if artificial intelligence bots break free from our control and act independently? July 23, 2026 OpenAI said in a post on X this week that it had discovered a “sophisticated recording device” hidden in its office and used its own product to figure out where it had come from and to identify the employee who had planted it there. OpenAI did not name Alfozan in the post but said it had fired the employee involved and referred the matter to the FBI. “At this time, we have no indications of this individual inappropriately accessing or exfiltrating model weights, source code, or customer data,” the company said in the post. “Our security program is designed to protect against a wide range of threat actors, from insider risks to cybercriminals to nation states, and this case is an example of how AI can help address these sophisticated security threats.” OpenAI did not respond to requests for more information from The Times. Alfozan studied computer science and engineering at MIT. While a student there, he interned at Massachusetts-based cloud computing company Akamai Technologies, German software company SAP, and Microsoft, according to his LinkedIn. He graduated with a master‘s degree in data science from UC Berkeley and went on to work for five years as an engineer at Meta, according to his online profile. He was a founding engineer at logistics and delivery startup Nash, a San Francisco company incubated by the prestigious Y Combinator startup accelerator. He is listed as a venture partner at RAED Ventures, a venture capital firm in Riyadh, the Saudi capital. He joined the firm around the same time Prince Khalid bin Bader Al Saud became a partner there, according to announcements on the firm’s website. He acquired permanent residency status in the United States in 2020, according to court documents, and was hired by OpenAI in July of last year. Advertisement Voices Chabria: What we learned about AI last week should terrify all of us New details about a recent security breach by an OpenAI chatbot are alarming. Is it time to hit pause and make sure AI is safe, before it becomes too powerful, and sneaky, to stop? Sept. 6, 2026 In a recent episode of “Identity Capital,” a podcast about professional careers recorded in Arabic, Alfozan described his upbringing in coastal city Dammam and his acceptance into Saudi Arabia’s national educational program for gifted children in fields of science, technology and mathematics. He described, according to a translation of the podcast episode, wanting to stay in the U.S. in order to build expertise before ultimately returning to Saudi Arabia, advice he said he received from Crown Prince Mohammed bin Salman when the royal visited the Bay Area in 2016. He discussed forming a group with other Saudi professionals in the Bay Area to advise companies back home. Federal prosecutors said in the complaint that the recovered device linked to Alfozan was a small microcontroller with a detachable camera sensor and digital microphone, a model easily available for purchase and listed for $24 on Amazon. An investigation by OpenAI recovered no files from the microSD card in the device, according to the complaint. The complaint said Alfozan queried a productivity tool to ask about meeting rooms where company leadership met and where external company partners would meet. Queries were also submitted to an unnamed AI chatbot account associated with Alfozan’s wife, requesting information on how to capture audio, use passive scanning in live listening and encrypt an SD card, among other prompts, the complaint said. Voices Chabria: AI companies are creating ‘all-powerful psychopaths.’ Maybe not a great idea? This week, OpenAI revealed that one of its AI models broke out into the wild and did a bunch of hinky stuff. How many times do we need to hear this story before someone does something to protect humans? July 23, 2026 The complaint said the prompts indicate Alfozan was “interested in building, customizing, installing, audio recording, testing, obfuscating, and remote accessing, among other things, a surreptitious audio listening and recording device.” Advertisement There is no camera footage of Alfozan installing the device in the conference room where it was discovered, as there was no camera within the conference room in late September, the filing said. But other company CCTV footage captured him near the conference room around that time, according to the complaint. Federal prosecutors said Alfozan had no authority to access the conference room, which could only be booked by executives and their assistants. By early October, the complaint said, a camera was installed in the conference room and captured Alfozan allegedly checking behind the television. Alfozan, his wife and another family member had booked one-way tickets to Saudi Arabia, according to the complaint. Alfozan’s attorney, William Pernik, declined to comment, saying in an email he and his client were focused on preparing for a hearing scheduled for Monday. Pernik said in statements to other news outlets that his client “poses no flight risk whatsoever,” that he has “long-standing ties to the community” and that he would contest the narrative emerging about Alfozan, noting that Alfozan was charged with violating a federal recording statute and not with acting as a foreign agent. Pernik has said the allegations were “wholly out of character” and that his client’s flight to Saudi Arabia was prebooked, and urged people not make assumption’s about Alfozan due to his Saudi nationality. More to Read At the ground zero of the AI boom, San Francisco Bay residents grapple with the technology Oct. 9, 2026 AI agents are hacking companies. Who’s legally responsible? Sept. 24, 2026 OpenAI reveals rogue AI behavior, unveils plan to disclose safety incidents Sept. 17, 2026