A new joint poll by Groundwork Collaborative and Ipsos finds that only about one-third of American workers expect artificial intelligence (AI) to make their jobs better. The poll’s results were first shared with Semafor.
Income and occupational gaps
The survey shows a clear income divide: 40% of respondents earning $100,000 or more said they expect AI to improve their job, compared with 19% of those earning under $50,000.
Occupational differences are also evident: white‑collar workers are more likely than blue‑collar workers to report using AI in their work.
Racial differences in perceived risk
The poll indicates racial variation in perceived risk from AI: Black workers are more likely than white workers to predict that AI will replace their jobs.
Reactions and policy proposals
Alex Jacquez of Groundwork Collaborative commented on the findings, saying, “This is really just yet another class divide for the American public.” Jacquez rejected proposals to address AI’s economic impacts with a universal basic income but said that “part of a solution” could be an idea advocated by Senator Bernie Sanders (I‑Vt.): a public stake in AI firms.
Why this matters
The poll highlights that the benefits and risks of AI are unevenly distributed across income levels, job types and racial groups. Those differences have implications for labor policy and technology regulation, because they can influence how economic inequality evolves as AI spreads.
Sources: joint public-opinion poll by Groundwork Collaborative and Ipsos; results shared first with Semafor.
