Americans increasingly see AI as potential threat to humanity, according to new poll

A majority of Americans believe advanced artificial intelligence carries at least a moderate risk of eventually causing the destruction of humanity, according to a Politico poll released Wednesday, reflecting growing public concern over the rapid pace of AI development.

Around two-thirds of respondents said there is at least a moderate possibility that advanced AI could destroy humanity. Of these, 17% considered the outcome “almost certain,” 20% described the risk as “significant,” while 26% rated it as moderate.

Concerns were reported across political, gender and age groups. Among voters in the 2024 presidential election, 70% of those who supported Democratic candidate Kamala Harris saw at least a moderate risk, compared with 60% of Donald Trump voters.

Trump has previously downplayed concerns surrounding AI.

The poll also showed stronger public support for slowing the development of AI than for continuing to advance increasingly sophisticated systems.

A new Politico poll has found that most Americans believe advanced artificial intelligence could pose a serious long-term threat to humanity, with many seeing at least a moderate possibility that it could eventually lead to humanity’s destruction.

About two-thirds of those surveyed said advanced AI presents at least a moderate risk. Around 17% considered such an outcome “almost certain,” while 20% called the risk “significant” and 26% described it as moderate.

The concern was shared across different political, age and gender groups. Among 2024 presidential voters, 70% of Kamala Harris supporters viewed AI as posing at least a moderate risk, compared with 60% of Donald Trump supporters. Trump has previously dismissed some concerns about the dangers of AI.

The survey also indicated that more Americans favour slowing the development of AI than continuing to build increasingly advanced systems.