OpenAI Is Warning Employees to Watch Out for Spies, Including Romantic "Honeypots," per WSJ
The friendliest stranger may be the threat
OpenAI is warning its employees to watch out for spies, especially romantic ones, per the Wall Street Journal.
The company has cautioned staff in internal trainings that someone trying to seduce them may likely be a "honeypot," an agent who uses the offer of sex or romance to get close to a target. Even strangers approaching them in everyday places could be gathering information.
In one incident described in the trainings, a stranger approached an OpenAI employee who was at a San Francisco playground with their child and steered the small talk toward the employee's work. Per the Journal, the stranger was a spy.
Champagne, a redhead, and a young man
To make the risk concrete, OpenAI tells employees the story of a foreign espionage operation against a US diplomat in Russia.
The operation first sent a blond woman to the diplomat's hotel room with Champagne. When he declined, the spies sent a redheaded woman, and then a young man.
The implication for OpenAI staff is that the same playbook could be run on them, according to people familiar with the trainings. Security staff also warn about bribery offers and eavesdropping in public places like cafes.
The day-to-day hygiene rules are just as strict. Employees are told not to wear OpenAI-branded clothing outside the office, to use privacy screens on phones and laptops in public transit, and to wipe whiteboards clean after meetings.
An arrest inside the lab
The warnings carry fresh urgency after what prosecutors describe as the first confirmed case of spying inside a frontier AI lab.
On October 9, the FBI arrested Abdulrahman Alfozan, a 35-year-old Saudi engineer who had worked at OpenAI since July 2025, after the company found a hidden recording device concealed in toilet paper behind a television in an executive conference room. A second similar device was later found in a bag he had left unattended. He faces up to five years in prison if convicted, and OpenAI has said there is no evidence he accessed model weights, source code or customer data.
The physical warnings sit on top of a broader security buildout the company accelerated after Chinese startup DeepSeek released a competing model in January. Proprietary technology is kept on offline computers, sensitive areas require fingerprint scans, and the network runs on a deny-by-default internet policy.
Anthropic and Google have reportedly given their employees similar warnings.
—
Do you want to see how to make more plays? Do you want to find gains yourself? Unusual Whales helps you find market opportunities through our market tide, historical options flow, GEX, and much, much more. Create a free account here to start conquering the market with Unusual Whales
—
Options market and stocks to watch
OpenAI itself is private, so there is no ticker to trade the headline directly. The exposure sits with its partners and rivals.
MSFT is OpenAI's biggest backer and the main public proxy for its technology, so any story about the security of OpenAI's research lands in Microsoft's narrative first.
NVDA supplies the compute the entire frontier AI race runs on, and espionage headlines underline why export controls and supply chain security keep showing up in its options flow.
GOOGL is in the same boat as a rival lab that has reportedly issued similar warnings to its own staff. Watch for flow in the AI complex if the story broadens into a sector-wide security theme, and follow developments on the Unusual Whales news feed.
Want more market intelligence? Create your free Unusual Whales account for options flow, market tide, GEX, and the full toolkit.