Americans Divided on Whether AI Will Make Things Better or Worse, New Survey Finds

WASHINGTON—Americans are closely split on whether they think artificial intelligence (AI) will make things better, worse, or have no impact on their own lives and careers. The country is also evenly divided on whether it will be good or bad for society and the next generation—but very few think it will have no impact, according to a new survey conducted by Public First in partnership with the Center for Data Innovation.

The nationally representative survey of over 2,000 U.S. adults was conducted earlier this month to track how public attitudes and perceptions of AI have shifted since Public First conducted a similar survey last year. The emotion Americans expressed most frequently about AI in both surveys is curiosity. But that has ticked downward since last year, while feelings of worry, anxiety, and fear have increased.

“In our polling, we saw that Americans had noticed the recent accelerating improvements in AI, but they were also deeply divided about whether we should be more excited or worried,” said Vinous Ali, Managing Director at Public First. “While some Americans were keen to experiment with AI tutors, coaches, or even relationships, others pointed to the threats from misinformation, automation and superintelligence. Getting the policy and regulatory environment is crucial but so too is building understanding of how to make best use of what AI could offer.”

Nearly 6 in 10 Americans (59 percent) think it is likely that AI will increase unemployment. Yet less than one-third of American workers (28 percent) say they have used an AI chatbot at work, and, among those who have, 68 percent say it has been helpful and 38 percent say it has become an essential tool they use regularly. Meanwhile, 70 percent of Americans would support using AI to diagnose health conditions if it’s double checked by a human, and 43 percent of American men under 35 even express interest in having an AI girlfriend or boyfriend.

Among the biggest concerns, 71 percent of Americans fear AI-generated content could manipulate elections (a concern shared almost equally among Democrats and Republicans).

When it comes to AI policy, 60 percent of Americans say government should move cautiously before creating new laws and regulations to avoid creating unintended consequences. When asked to choose between two policy priorities—either staying at the technological frontier or developing AI responsibly, even if it allows countries like China taking the lead—opinions are divided, with 43 percent of Americans preferring to stay at the frontier and 35 percent preferring responsible development.

“Policymakers have their work cut out for them, not only to design the right policies for AI and keep the United States on course in the global AI race, but also to build public support for these initiatives,” said Daniel Castro, director of the Center for Data Innovation. “This task is especially challenging because AI and the companies that make it are often vilified in the media. As more people use AI, experience the benefits, and realize their worst fears have not come to pass, hopefully public support will follow.”

The Center and Public First are releasing the new survey findings at a briefing event that will be livestreamed from Capitol Hill in Washington, DC, Tuesday at noon Eastern. More findings follow.

Today:

• On the question of whether AI will make things better or worse for Americans personally, 32 percent of U.S. adults say it will make things better, 22 percent say it will make things worse, and 33 percent say it will have no impact.
• When the question turns to whether AI will be good or bad for society as a whole, 38 percent of Americans say it will make things better, 35 percent say it will make things worse, and just 9 percent say it will have no impact.
• A 39 percent plurality of American say they are curious about AI in this year’s survey, down from 42 percent last year.
• Among negative emotions, 37 percent say they are worried about AI (up from 32 percent in 2023); 28 percent say they are anxious (up from 23 percent); and 23 percent say they are scared (up from 19 percent).

Opportunities:

1. Large majorities of Americans express interest in using AI for a wide range of personal purposes, from learning new information (73 percent) to helping reduce unnecessary energy usage (66 percent), to managing health and fitness (60 percent).
2. Majorities of Americans under 35 express also interest in an AI personal assistant (62 percent), tutor (51 percent), workout coach (51 percent), or financial advisor (51 percent).

Concerns:

1. 62 percent of Americans are worried about the potential impact of misinformation on the U.S. election, and 46 percent lack confidence in their ability to detect fake AI-generated content.
2. 59 percent believe AI will likely increase unemployment, and just 7 percent think it is likely to increase workers’ wages.
3. Those with a bachelor’s or master’s degree are slightly more likely to believe AI could replace their jobs within 20 years.

Policy:

1. Majorities of Americans support a wide range of AI-related policies on domestic issues, from clear labeling of AI-generated content (66 percent) to banning the use of AI-generated content for political campaigns (57 percent), to holding AI companies legally liable (64 percent).
2. By contrast, policies which more Americans saw as a bad idea than a good idea were introducing new taxes on AI models (32 percent opposed, 24 percent supported), banning new research into AI (35 percent opposed, 22 percent supported) and increasing government funding for AI research (34 percent opposed, 22 percent supported).

“The challenge that policymakers now face is perfectly encapsulated in the fact that public currently opposes more government funding for AI research, which is at the top of the policy agenda that the Bipartisan Senate AI Working Group has just released,” said Castro. “To remain a global leader in AI, the United States needs to stay on the frontier of AI development, which requires maintaining a robust ecosystem of AI skills, chips, data centers, and models. It also needs to lead in AI adoption, especially in traded sectors of the economy where it faces global competition, and areas like education and government where there are significant rewards. Both efforts will require extensive coordination and cooperation between the public and private sectors, as well as a regulatory environment that fosters responsible innovation.”

1. Read the full report. (PDF)
2. Register for the briefing event.

Contact: Nicole Hinojosa, [email protected].