Americans worry AI companies aren't doing enough to prevent harm

A new Reuters/Ipsos poll finds most Americans are concerned that AI companies are not doing enough to prevent serious harm, with 39% now viewing AI as having a negative impact on society, the highest share on record. The poll, conducted over four days ending Sunday, shows 39% of respondents see AI as harmful, up from 36% last month, while only 11% view it positively. Additionally, 73% prioritize safe and responsible AI development over maintaining U.S. dominance, and 55% support slowing AI development. The findings reflect broader skepticism, with 69% aware of AI companies advocating for reduced development pace. Federal AI legislation remains limited, as the Republican-controlled House is in recess through November, and separate Ipsos data shows two-thirds of Americans think AI development is too fast.