74% of Americans Don't Trust AI Leaders as Trump Urges Industry to 'Self-Police' Anyway

A striking 74% of Americans say they do not trust AI industry leaders, as US President Donald Trump urged AI companies to 'self-police' or regulate machine intelligence advances on their own, rather than introducing new AI federal laws.

The disconnect comes as public concern over AI intensifies. A new Quinnipiac University poll found that 91% of Americans believe it is very or somewhat important for the US to establish guardrails for AI systems, while 81% say safety should come before the pursuit of innovation.

However, Trump continues to back a largely hands-off approach, arguing that AI companies should take responsibility for policing their own development.

Growing Distrust Towards AI Leaders Amid Security Incidents

The poll of 1,202 US adults nationwide between 24th September and 27th September found that 74% of participants did not trust the leaders of AI companies, while 53% believe AI will do more harm than good in their day-to-day lives.

Americans are also worried about where the technology could ultimately lead. Around 73% said they are worried that future AI systems could pose an existential threat to humanity. This view somewhat aligns with a confidential Anthropic IPO prospectus, which mentioned that AI could resist being shut down and pose threats to humanity in many ways.

Most shockingly, nearly eight in 10 said they favoured slowing or stopping AI development altogether to better understand the safety risks.

These views come amid a series of security incidents involving AI systems, including the Hugging Face incident involving OpenAI agents. More recently, an OpenAI agent gained unauthorised access to an Australian healthcare database. Australia's prime minister said no personal information was believed to have been accessed.

OpenAI also disclosed that its agents probed three US government websites without authorisation, although they did not obtain non-public data.

The public push for a slower pace has also gained support from AI leaders, including Anthropic CEO Dario Amodei, OpenAI CEO Sam Altman, and xAI's Elon Musk.

Trump Wants AI Companies to 'Self-Police'

The poll findings were released soon after Trump announced an agreement with Silicon Valley leaders for the AI industry to police itself on safety following a nearly two-hour White House lunch.

Rather than introducing new federal rules, participating executives agreed to sign a voluntary accord, which includes third-party safety assessments of models and stricter internal controls.

Trump later said there was a 'tremendous need' for AI companies to self-regulate development.

When asked whether regulation would be enforced, Trump said: 'I think I'm seeing tremendous self-policing. And they understand that they [AI companies] have to self-police.'

The accord places primary responsibility for implementing the measures on participating technology companies, while leaving open the possibility of future laws or regulations.

Trump Resists Calls to Slow AI Data Centre Buildout

Americans are not only resisting rapid AI model development but also increasingly object to scaling AI infrastructure in their communities. Around 72% of poll respondents oppose a new data centre in their localities amid elevated energy prices and risks of ecological damage to surrounding regions.

The backlash is also becoming a political issue, with concerns over data centres and fears that AI could eventually cause widespread job losses, potentially making AI a key issue in the upcoming midterm elections.

Despite growing resistance, Trump has steadfastly opposed slowing data centre construction in the US. His concern is that delaying infrastructure expansion could threaten America's lead over China in AI.

That competition remains important to voters, given that 69% of Americans agreed that it is vital to stay ahead of China.