• OpenAI says it discovered a sophisticated recording device secretly planted in one of its offices.
  • The company used its own AI models to investigate and identified an employee who allegedly installed it days earlier.
  • The employee was terminated and the FBI was notified; OpenAI says its investigation has not identified any unauthorized access to model weights, source code, or customer data.

OpenAI has notified the FBI after discovering a sophisticated recording device secretly planted inside one of its offices, according to a company statement. The artificial-intelligence developer said it used its own AI models to investigate the device and identified an employee who allegedly installed it days earlier. The employee has been terminated.

The company said its investigation has not identified any evidence of unauthorized access to or theft of model weights, source code, or customer data. “We have notified the FBI and are cooperating fully with their inquiries,” an OpenAI spokesperson said. The office location, discovery date, device capabilities, and employee identity remain unknown. The FBI declined to comment on whether it has opened an investigation or brought charges.

Notifying the FBI Is Not the Same as an Investigation

A critical distinction: notifying the FBI does not establish that the bureau has opened a formal investigation or filed charges. Similarly, “no evidence of exfiltration at this time” is a provisional finding, not proof that nothing was recorded or transmitted. Even without theft of digital files, the alleged capture of confidential conversations would be a separate issue. OpenAI said its investigation is ongoing.

The incident comes as OpenAI faces heightened scrutiny over its security and governance practices. In March 2026, the company reported more than 900 million weekly active ChatGPT users and more than 50 million subscribers. CNBC (VSNT) reported on October 13 that OpenAI had shared an approximately $50 billion annualized revenue figure with investors, though earlier reports cited roughly $70 billion. The lower figure excludes cloud-partner revenue, according to a person familiar with the matter. OpenAI announced $122 billion in committed capital at an $852 billion post-money valuation on March 31, 2026, and Reuters (TRI) reported on September 29 that the company was seeking at least $30 billion at an approximately $1.4 trillion pre-money valuation. Those discussions were preliminary.

A Pattern of Security Concerns

The alleged device incident adds to a series of security-related developments at OpenAI. In July 2024, reporting revealed that a hacker stole information from internal employee discussions in 2023, though did not access the systems used to build and house OpenAI’s AI. A September 29, 2026, New York Times (NYT) report described employee and researcher warnings about OpenAI’s testing safeguards and corporate infrastructure. And on September 25, the BBC reported that OpenAI notified dozens of institutions about improper activity by its agents, including government websites.

The alleged use of AI to identify an employee raises unresolved questions about what evidence supported the identification. The supplied account does not describe forensic corroboration, human review, or the model’s role. OpenAI declined to provide further details.

Legal and National-Security Stakes

U.S. federal law generally prohibits intentional interception of protected oral, wire, and electronic communications, subject to statutory exceptions. Whether this alleged recording violated the law would depend on the circumstances, including consent and the nature of the communications. A relevant precedent is former Google (GOOGL) engineer Linwei Ding’s January 2026 conviction for economic espionage and theft of confidential AI technology. Prosecutors said he stole more than 2,000 pages for the benefit of China. That case demonstrates the national-security stakes of AI intellectual property, but supplies no evidence of foreign involvement here. The supplied account does not identify a foreign government, competitor, or recipient.

OpenAI’s government work has separately sparked debate over surveillance safeguards. The company says its defense agreement prohibits domestic surveillance; the Electronic Frontier Foundation has criticized the agreement’s protections. This is relevant political background, not evidence about the alleged device.

What Comes Next

In the short term, confirmation would likely prompt forensic examination of the device, preservation of access records, review of potentially captured conversations, and questions from employees and enterprise customers. The most consequential findings would be whether the device actually recorded, whether it transmitted anything, and who could access the recordings. Over the longer term, consequences could range from a contained personnel matter to criminal proceedings, privacy claims, or broader security reforms.

The immediate information gap is the original report or OpenAI statement. Without it, the alleged device incident cannot responsibly be presented as confirmed news, even though the surrounding financial, governance, and security developments are documented. OpenAI says its investigation is ongoing. The company did not respond to a request for additional comment.