Tech giants split over calls to slow down AI as poll finds 63% of Americans fear it could 'destroy humanity'

From the Newshound archive · first published 19 September 2026.

A new poll shows that close to two-thirds of Americans think AI poses an existential threat to humanity — a striking turnaround from just months earlier, when the public was evenly divided on the question. Anthropic CEO Dario Amodei has called on the AI industry to pump the brakes on development, but OpenAI and Meta are pushing back, laying bare a sharp divide among Silicon Valley's most influential companies. Elon Musk has similarly advocated for a slowdown, drawing accusations from critics that he is simply trying to hobble competitors. Top labs are cautioning that AI systems able to autonomously enhance themselves may emerge sooner than previously thought.

Bernie Sanders and Steve Bannon found rare common ground at an anti-AI conference in Washington this week.

Christine Lagarde warned Europe faces a stark choice: shun AI and lose growth, or embrace tools built by rivals in the US and China.

Critics argue the existential doom warnings are a 'sleight of hand' to deflect scrutiny over AI's real, immediate harms — from dangerous companion apps to environmental damage from datacentres.