OpenAI and Anthropic are reportedly tightening security precautions for employees in San Francisco, advising staff to conceal company branding in public amid concerns about espionage, attempts to obtain confidential information and growing hostility towards the artificial intelligence industry.
The measures reflect the sensitivity surrounding companies developing advanced AI systems.
Employees may have access to commercially valuable research and corporate information, making their professional connections potentially attractive to outsiders seeking insights that are not publicly available.
OpenAI and Anthropic Tighten AI Employee Security
OpenAI reportedly advises staff against wearing clothing displaying its logo outside the workplace. Its security training also warns employees about approaches that could be intended to extract confidential information, including offers of money, friendship or romantic interest.
Anthropic has adopted a similarly restrictive approach to company merchandise. The New York Post reported on 26 October 2026 that the Claude developer had largely banned branded items for employees and reminded staff to conceal their identification badges when leaving its premises.
It’s a cultural shift!
— Vivi (@vivilinsv) October 26, 2026
Company swag is still a badge of pride at places like Google and Apple. But at OpenAI and Anthropic, employees are being told to keep company logos out of sight outside the office amid concerns about espionage, seduction and physical attacks.
And here’s a… https://t.co/Bj3J5homx6
Anthropic is also concerned about people photographing its offices, while OpenAI reportedly advises staff to use privacy screens on public transport and avoid discussing sensitive work in public.
Miao Hong, managing partner of Silicon Valley Future Capital and an investor in both companies, told the Post that outsiders seek insider knowledge of cutting-edge technology, making employees cautious about what they disclose.
An unfamiliar person asking questions about AI is not necessarily acting maliciously, and ordinary curiosity should not be confused with espionage. The challenge for security teams is to recognise potentially suspicious approaches without treating every social interaction as a threat.
OpenAI Recording Device Case Raises Security Concerns
On 9 October, Abdulrahman Alfozan, a 35-year-old Saudi national who worked as an OpenAI engineer, was arrested and charged in connection with an alleged attempt to intercept private conversations at the company's San Francisco offices.
A federal criminal complaint described a recording device concealed behind a television in an executive conference room, according to reporting on the case. The device was reportedly wrapped in toilet paper and inserted into the television's USB port.
OpenAI dismissed Alfozan following the incident. The case subsequently progressed beyond the initial arrest, with a federal grand jury indictment reported on 28 October.
The allegations against Alfozan remain unproven, with no confirmed foreign-government involvement. Separately, a reported encounter at a San Francisco playground raised suspicions of an attempt to obtain information from an OpenAI employee, though its significance remains unclear.
AI Workers Face Pressure Beyond Espionage
AI companies also face public backlash over the technology's social impact, environmental costs and military applications. In July, Corgi chief executive Nico Laqua told the Wall Street Journal that passers-by shouted abuse about AI outside the startup's San Francisco cafe, citing rising rents and water consumption.
OpenAI has also faced protests over its Pentagon work. In April 2026, an alleged firebombing and a separate shooting investigation targeted Sam Altman's home.
One suspect was arrested over the firebombing, while two people arrested after the shooting were released as prosecutors reviewed the evidence.
Anthropic reported spending $2.75 million (£2.06 million) on chief executive Dario Amodei's personal security in 2025, according to reporting on its regulatory filing.
The security concerns are also affecting employees' personal lives. The father of an Anthropic worker told the Post that his son and colleagues avoid discussing work socially, while finding it harder to know whom to trust. He said people often focus on their wealth and employer rather than their character.
