Economists expect AI to reshuffle white-collar work, not erase it, new survey finds

Indeed Hiring Lab and Pulsenomics surveyed 120 economists and labor market experts in July 2026 and found a labor market that has already cooled, rather than one still cooling. Unemployment is expected to rise only modestly, from 4.2% to roughly 4.4% by year end, and Indeed's own Job Postings Index is projected to fall 1.4% by June 2027.

The panel split on how AI factors in. 52% expect AI to hurt employment over the next year, 35% expect it to help, and 13% see no effect. The more consistent signal is on pay: 57% expect AI to put downward pressure on wages for college-educated workers, compared with 34% for workers without a degree.

The sharpest disagreement is over where the cuts land. Software development and administrative support are the roles economists flagged for the steepest declines, while hands-on healthcare work, personal care, home health, and nursing, is expected to keep growing because it sits largely outside AI's reach. The framing isn't developers being replaced wholesale, it's reallocation within white-collar and technical work toward roles that complement AI tools rather than compete with them.

For a small team hiring engineers, or a developer deciding how to position their own role, that reallocation is the practical signal: junior generalist software work is what economists expect to shrink first, not senior or AI-fluent roles.