Discussion of AI and jobs tends to swing between two extremes — mass unemployment or business as usual. The reality happening in the labor market right now is more specific: AI is changing which tasks within a job get automated and which skills become more valuable, rather than eliminating entire professions wholesale.
Tasks, not jobs, are the unit of change
Most jobs involve a mix of tasks with very different automation potential. A writer’s job includes drafting, editing, fact-checking, and pitching ideas — AI is currently strong at drafting assistance but much weaker at original reporting or building relationships with sources. This means AI often changes how a job is done rather than eliminating it outright, at least for now.
Skills gaining value
The ability to direct, evaluate, and edit AI output — sometimes called “AI literacy” — is becoming a valuable skill across many fields. Judgment, domain expertise, and the ability to catch when AI output is wrong are increasingly what separates effective AI-assisted workers from those who simply take AI output at face value.
Where genuine disruption is more likely
Roles built almost entirely around highly routine, well-documented tasks — certain types of data entry, basic transcription, simple content moderation — face more direct pressure than roles requiring judgment, relationship-building, or handling genuinely novel situations. How this plays out varies significantly by industry and is still actively unfolding, which is why economists and labor researchers continue to track it closely rather than treating it as a settled question.