US concern over AI has climbed as adoption expands, Pew data show

American concern about artificial intelligence has risen sharply since 2021, even as consumer use of chatbots and other AI tools has spread, according to Pew Research Center findings cited by Ngaire Woods in Project Syndicate. Pew found that the share of Americans who were more worried than enthusiastic about AI increased from 37% in 2021 to 52% in 2023.

Woods, writing from Oxford, argued that the technology industry is seeking wider public acceptance of AI at a time when confidence in government is weakening. She said those issues are linked because people are less likely to trust AI systems if they do not trust the public institutions charged with supervising them.

Pew’s more recent research, cited by Woods, found that more US adults now expect AI to have a negative rather than positive effect on both their personal lives and society. The available figures do not provide a detailed breakdown of the latest margin, but they point to a shift in public sentiment as AI moves further into daily use.

Trust becomes part of the adoption equation

The mechanism is straightforward: consumers and workers are being asked to rely on tools whose outputs can affect information, jobs, services and decision-making. If users believe that regulators lack credibility or capacity, industry assurances may carry less weight.

Woods framed AI acceptance as more than a technical challenge. In her account, the task for the technology sector is not only to make AI tools useful or accessible, but also to persuade the public that the systems operate within rules people regard as legitimate.

That argument places public institutions at the center of AI adoption. Regulators do not build most commercial AI products, but they help set expectations for safety, accountability and redress. Woods’ point is that public skepticism toward those institutions can spill over into skepticism toward the technology itself.

Pew surveys show widening caution

The Pew data cited by Woods show concern rising over a two-year period in which AI tools became more visible to the public. In 2021, a minority of Americans told Pew they were more concerned than excited about AI. By 2023, that group had become a majority.

Woods also said public anxiety has been increasing faster than chatbot adoption. That comparison suggests that greater exposure to AI tools has not, on its own, resolved doubts about the technology’s broader consequences.

The findings matter for companies, policymakers and public agencies seeking to expand the use of AI in workplaces and public services. The data indicate that adoption may depend not only on product performance, but also on whether Americans believe the institutions overseeing AI can be trusted to manage its risks.

This story draws on original reporting from Project Syndicate.