AI Lab Staff Hide Badges as Threats Mount
OpenAI and Anthropic security measures reflect espionage concerns and violent incidents tied to AI work, a trend without a settled cause.
Updated on Nov. 3, 2026 in Artificial Intelligence
Live Poll
Do you trust major AI companies to keep their employees and the public safe?
OpenAI and Anthropic have instructed employees to conceal company badges and logos in public, and OpenAI has issued security training warning staff about espionage approaches involving money, friendship, or romance. The measures are operational precautions already in place, not announced products or research.
Why it matters
The precautions reflect concern about espionage, unwanted approaches, and hostility toward people working in artificial intelligence, a field whose leading labs now treat staff identity as a security surface. The shift follows protests at OpenAI's San Francisco offices and violent incidents at its CEO's home.
The disclosed figure is a single $2.7 million expenditure on personal security services for Anthropic CEO Dario Amodei, reported in the company's listing filing. The source provides no baseline for prior security spending, and the motives behind the incidents remain unestablished pending trial.
The players
OpenAI
San Francisco-based artificial intelligence developer whose models underpin ChatGPT and whose staff security training is described in this report.
Anthropic
San Francisco-based artificial intelligence safety company, developer of the Claude model family, which disclosed security spending in its stock-market listing filing.
Sam Altman
Chief executive of OpenAI, whose home was targeted in two incidents in April 2026.
Dario Amodei
Chief executive of Anthropic, for whom the company disclosed $2.7 million in personal security spending.
The details
OpenAI's security training describes approaches ranging from overhearing conversations in cafes to bribery and attempts to build personal relationships with staff to extract confidential information. Employees are told to use privacy screens on public transport and to erase meeting-room whiteboards before leaving. Anthropic has reminded workers to hide badges and has largely restricted branded merchandise, and has reported people trying to photograph its offices.
Timeline
March 2026: Activists wrote messages outside OpenAI's San Francisco offices urging staff to quit over its Pentagon agreement.
April 2026: OpenAI CEO Sam Altman's home was targeted in two separate incidents.
Two days after the first April incident: Two suspects were accused of firing a shot outside Altman's home.
The Tech Race
The protests over OpenAI's Pentagon agreement in March 2026 preceded the April incidents at Sam Altman's home, and the security steps now reach individual employees and executives. The sequence shows how defense-related AI contracts are drawing scrutiny directly onto the people who build these systems.
Employees at AI labs now face explicit guidance on hiding badges, limiting branded merchandise, and guarding conversations in public, which changes daily routines for staff. The source does not indicate whether these precautions extend beyond current employees or to the general public.
The takeaway
Watch the trial outcomes for the three arrested suspects, who have pleaded not guilty, for any indication of motive tied to AI work. Track whether OpenAI or Anthropic publish further security disclosures in later listing or regulatory filings.
Further reading
For more on the labs at the center of this story, see Artificial Intelligence coverage from San Francisco.
Source note: This article includes information reported by The Times of India.
Live Poll
Do you trust major AI companies to keep their employees and the public safe?
