In a sign of escalating real-world tensions surrounding artificial intelligence, workers at leading AI labs, including Anthropic and OpenAI, are reportedly being advised to hide their corporate affiliation in public.
A New Kind of Workplace Hazard
Reports emerging in late 2026 suggest
that top AI companies are implementing new security measures for their staff that go beyond typical corporate confidentiality. Anthropic, an AI safety and research company, has reportedly advised its employees to conceal their identification badges when outside the office and has limited the availability of company-branded merchandise. This guidance is not unique to Anthropic. OpenAI, another major player in the AI race, has reportedly given its staff similar instructions, telling them not to wear items with the company logo in public. This shift marks a departure from the traditional Silicon Valley culture where company 'swag' is often worn as a badge of honour. The new policies reflect a growing concern within these influential labs that their employees could become targets.
Espionage, Hostility, and Physical Threats
The security concerns are multifaceted, ranging from corporate espionage to direct physical hostility. OpenAI's security training reportedly warns employees about various tactics used to extract confidential information, from strangers striking up conversations in public places to more elaborate attempts involving offers of money or building personal relationships. An investor in both companies noted that employees are constantly approached by people seeking inside information on the cutting edge of AI development. Beyond corporate intelligence gathering, there is a backdrop of increasing public animosity. Protests have become a regular occurrence outside the San Francisco offices of Anthropic and OpenAI, with demonstrators calling for a halt to the 'AI race'. These concerns escalated dramatically in 2026, with reports of physical attacks targeting the home of OpenAI CEO Sam Altman, including an incident involving a Molotov cocktail.
The High Cost of Security
The rising tide of threats is forcing AI companies to invest heavily in protecting their people and intellectual property. According to one report on its regulatory filings, Anthropic spent a significant sum on personal security for its CEO, Dario Amodei. The security measures extend to daily routines, with employees at firms like OpenAI being reminded to use privacy screens on their devices while on public transport and to erase whiteboards after meetings. The climate of fear has also impacted the personal lives of employees and their families. The father of an Anthropic worker described how his son and colleagues now avoid discussing their work in social settings and struggle to trust new acquaintances, unsure if the interest is genuine or aimed at their employer.
An Industry on the Defensive
These new personal security protocols reflect a broader, more defensive posture for an industry grappling with its own creations. Public discourse is filled with anxieties about AI's potential, from job losses and misinformation to more extreme 'existential risks'. Ironically, companies like Anthropic, which was founded with a focus on AI safety, are now dealing with very tangible, human safety problems. The abstract fears about what AI could become are manifesting as concrete threats against the people building it. The need for employees to hide their identities illustrates a profound and troubling disconnect between the creators of this world-changing technology and a public that is growing increasingly wary of its power and the speed of its development. The measures taken by Anthropic and OpenAI are more commonly associated with defense contractors or classified government agencies, signaling a new, more contentious chapter for Silicon Valley.
