Most Americans Consider AI a Threat to Humanity, New Poll Finds

It turns out that years of movies showing countless doomsday scenarios around advanced AI does not endear the public to the idea of mainstream adoption of artificial intelligence. A new Reuters/Ipsos poll finds that 61% of Americans see A.I. as a threat to humanity—22% did not, and 17% were unsure.

Since OpenAI’s ChatGPT was released in November, the race to bring generative A.I. to the public went into overdrive as companies like OpenAI, Google, and Microsoft competed for market dominance. Even though Skynet and the T-800 are still works of fiction, billionaire investor Warren Buffett likened the A.I. arms race to the development of the atomic bomb.

“It can do all kinds of things and when something can do all kinds of things I get a little bit worried,” Buffett said earlier this month. “I know we won’t be able to uninvent it and, you know, we did invent—for very, very good reason—the atom bomb.”

Buffett is not alone in issuing warnings about A.I. OpenAI co-founders Elon Musk and Sam Altman have also voiced concerns about potential dangers.

“Even benign dependency on A.I./automation is dangerous to civilization if taken so far that we eventually forget how the machines work,” Musk tweeted.

Musk’s concern around A.I. even led to the Tesla and Twitter CEO signing an opening letter calling for a pause in AI development in March—although he later acknowledged he knew it wouldn’t happen.

A more tangible view of the threat A.I. poses is emerging in the workforce. The striking Writers Guild of America included a proposal to block the use of artificial intelligence in TV and film production as part of their negotiations with the Alliance of Motion Picture and Television Producers.

Reuters says the online poll, taken by 4,415 adults in the U.S. and conducted between May 9 and May 15, shows that distrust in A.I. skewed higher amongst Americans who voted for former President Donald Trump at 70%, compared to 60% who voted for current President Joe Biden. Reuters did not say how those numbers were reflected across metrics like age and gender.

During a Senate Judiciary Committee hearing on Tuesday, Altman called for the regulation of A.I. to ensure that safety standards regarding technology development are followed.

“I think we also will need rules, guidelines about what’s expected in terms of disclosure, from a company providing a model that could have these sorts of abilities that you talk about, so I’m nervous about it.” Altman said. “I think is going to require a combination of companies doing the right thing, regulation and public education.”