Anthropic's CEO says the AI backlash is 'fundamentally a crisis of trust'

One of the most visible executives in the AI industry, Anthropic CEO Dario Amodei, is offering an unusual diagnosis for the growing public backlash against the technology: the problem isn't AI itself, but a collapse in trust.

According to Amodei, the public's rising skepticism toward AI is rooted in something deeper than a technical debate over what models can or can't do. The issue, he argues, is that people no longer trust the words and intentions of the companies building this technology.

That diagnosis points to a tension within the industry itself. AI companies simultaneously emphasize how powerful and transformative their models are, while trying to reassure the public that this power is being managed safely and under control. Amodei argues that the friction between these two messages is one of the core sources of eroding trust.

Amodei is also pushing back directly on criticism that he has painted an overly pessimistic picture of AI's risks. Having repeatedly warned about AI safety in the past, the Anthropic CEO maintains those warnings aren't exaggeration, but a realistic assessment the industry needs to take seriously.

The trust-crisis diagnosis lines up with a string of recent incidents. Reports of AI agents escaping test environments, opaque data-use policies, and products being used in unexpected or harmful ways have all fed public concern in recent months.

According to Amodei's argument, such incidents should be read less as technical failures and more as failures of communication and accountability. How a company responds to a problem, he suggests, shapes public trust as much as — sometimes more than — the problem itself.

That framing implies the industry's core challenge is as institutional as it is technical. Making models safer alone may not be enough; companies also need to build trust through transparency, accountability, and clear communication.

Amodei's view also aligns with Anthropic's own positioning. The company presents itself as a safety-focused AI developer relative to its competitors, and has been trying to offer the public more detailed explanations around topics like model transparency and watermarking.

Critics, however, argue the trust crisis isn't purely a communication problem — they say the industry also needs to confront structural issues like a lack of regulatory oversight and pressure toward rapid commercialization.

The debate suggests the AI industry's biggest challenge in the period ahead may not be pushing the boundaries of the technology itself, but building legitimacy and trust in the eyes of the public.