Interview, Podcast
How Will AI Impact the Labor Market?
- Joseph Briggs forecasts a monthly drag on job growth of 10,000 to 15,000 positions across tech, management consulting, and graphic design sectors.
- Approximately 9% of the U.S. workforce, equating to roughly 15 million workers, is projected to be reallocated to new positions over a 10-year transition period.
- If the 9% displacement occurs over a decade, the annual increase in the unemployment rate is expected to remain below one percentage point.
- A 5% acceleration in the pace of new job creation is anticipated to be sufficient to reabsorb workers displaced by AI automation.
- Historical trends suggest that approximately 85% of job growth over the last 80 years was driven by technological creation of new positions, supporting the expectation of net job gains.
- The adoption of AI is expected to be uneven and slower than capability improvements, with large businesses automating before small businesses due to the long tail of adoption.
- Labor outcomes will vary based on whether expert or inexpert tasks are automated, potentially leading to increased wages for fewer expert workers or decreased wages for more inexpert workers.
- Significant unemployment is considered premature to predict, with the impact characterized more as a manageable "rising tide" than a sudden shock.
- Darren Acemoglu predicts limited net job losses of 2% to 4% within the next five years, contingent on current AI investment trends.
- Net job losses are expected to be larger within the next 10 to 15 years if AI investment continues to focus on replacing workers rather than complementing them.
- By 2027, there may be increased layoffs or hiring slowdowns in roles where AI can replace specific tasks.
- The integration of AI and robotics is identified as a major variable that could significantly expand the number of impacted jobs.
- Future impacts are expected to shift toward jobs involving judgment, specifically middle management, and those requiring social interaction, which need significant AI model improvements and consumer preference changes.
- Increased labor income inequality is expected as some groups face stagnant wages and slower employment growth, rather than a decline via the replacement of high-paid managers.
- AI currently lacks the capability to fully complement workers outside of specific fields like biological or chemical research.
- While AI capabilities are improving rapidly, the transition from capability to job change involves multiple hurdles including cost-effectiveness and economic attractiveness.
- If the pace of job creation matches historical averages since the 1970s despite destruction in cognitive office jobs, the resulting net losses will remain limited.
- Uncertainty remains regarding whether future job creation will match job destruction, as there is no general economic law guaranteeing this balance.