Stanford’s Digital Economy Lab released an updated research paper analyzing the effects of artificial intelligence on the labor market. Using ADP payroll data through August 2026, the report highlights a significant shift in hiring patterns for early-career workers.
Employment for individuals between the ages of 22 and 25 in occupations with high AI exposure is now 19% lower than it would be if they followed the same trend as workers in less-exposed roles. This number rose from 15% just one year ago. The data indicates that firms are hiring fewer entry-level workers in these specific fields rather than firing current staff.
The decline centers on roles that rely heavily on codified knowledge that automated systems reproduce with increasing accuracy. Conversely, more experienced staff members in the same occupations remain stable. Their value remains tied to tacit knowledge gained through years of judgment and on-the-job experience.
This trend creates a clear challenge for workforce planning. If artificial intelligence takes over entry-level tasks, organizations must figure out how to provide the training ground necessary for junior employees to develop their professional judgment. Without a new strategy to replace these formative tasks, the long-term talent pipeline faces a potential disruption.