AI and the Labor Market: Early Signals, Not a Full Disruption

Artificial intelligence has rapidly moved from a futuristic concept to a mainstream business tool. Today, a large majority of U.S. companies report using AI in some form, most commonly for text generation, research assistance, coding support, and workflow automation. Yet despite the widespread adoption of AI tools, the economic impact on labor markets remains limited. Most businesses indicate that AI has not yet materially changed their operations, productivity levels, or workforce requirements. However, expectations for the coming years are considerably different, with many firms anticipating that AI will eventually boost productivity while reducing demand for certain categories of labor.

The challenge for investors, employers, and policymakers is distinguishing between current realities and future expectations. While headlines frequently suggest that AI is already replacing workers at scale, labor-market data paints a more nuanced picture. Employment levels remain relatively resilient, unemployment rates remain historically moderate, and many industries continue to face labor shortages. At the same time, emerging evidence suggests that AI may be beginning to influence hiring decisions and workforce composition, particularly in entry-level knowledge-based occupations.

History offers a useful framework for understanding how technological innovation affects employment. New technologies rarely eliminate work altogether. Instead, they tend to reshape job composition while creating entirely new categories of employment. Ecommerce provides a clear example. While online shopping disrupted portions of traditional retail employment, it simultaneously generated substantial demand for transportation, logistics, warehousing, and fulfillment workers. As consumers increasingly prioritized convenience and rapid delivery, employment in transportation and warehousing expanded dramatically.

The chart above illustrates this phenomenon. Since 2000, employment in transportation and warehousing has significantly outpaced both overall employment growth and employment growth in retail trade. The divergence became particularly pronounced following the acceleration of ecommerce adoption and intensified further during the pandemic era. While retail employment remained relatively stable over the long term, transportation and warehousing experienced substantial expansion as supply chains evolved to support new consumer behaviors.

AI is likely to follow a similar pattern. Rather than simply replacing workers, it is expected to alter the mix of jobs across the economy. Some tasks may become automated, others may be augmented, and entirely new occupations may emerge. The key question is not whether jobs disappear, but which jobs grow, which jobs shrink, and where economic value becomes concentrated.

Recent research examining AI exposure across occupations reinforces this perspective. Theoretical AI capability is significantly broader than actual AI usage today. While large language models can technically perform portions of many occupations, real-world implementation remains limited. Current adoption is concentrated in specific tasks such as coding, customer service support, content generation, data analysis, and administrative functions. Physical occupations, skilled trades, healthcare procedures, hospitality roles, and many service-sector jobs remain far less exposed to AI-driven automation.

Importantly, occupations with higher levels of AI exposure are projected to experience somewhat slower employment growth over the next decade. These occupations also tend to employ workers who are more educated, higher paid, and more likely to work in office-based environments. Yet despite these characteristics, there is currently little evidence that AI has caused widespread unemployment among highly exposed workers.

The labor-market data provides important context. Rising unemployment among recent college graduates has often been cited as evidence that AI is already replacing entry-level white-collar workers. However, the trend predates the emergence of generative AI by several years. As the chart above demonstrates, unemployment among recent graduates began increasing well before large-scale deployment of AI tools in late 2022.

This distinction matters because it highlights the role of broader economic forces. Elevated interest rates, slower hiring activity, geopolitical uncertainty, and a maturing economic cycle have all contributed to softer labor-market conditions. The current environment is frequently described as "slow-to-hire and slow-to-fire," where employers remain cautious about adding staff but are also reluctant to conduct large-scale layoffs. These dynamics affect recent graduates disproportionately because they typically rely on new hiring rather than existing employment relationships.

Nevertheless, some evidence suggests AI may be influencing labor markets at the margins. Recent studies indicate that employment growth among younger workers in highly AI-exposed occupations has lagged behind less-exposed occupations since 2022. Researchers have also observed signs of slower hiring activity in occupations where AI can perform a meaningful share of entry-level tasks. Rather than causing widespread layoffs, AI may initially reduce demand for new hires by enabling existing employees to handle larger workloads more efficiently.

This distinction between hiring and firing is critical. Labor-market disruption often begins with fewer job openings rather than immediate job losses. Entry-level positions are particularly vulnerable because many involve routine tasks that are relatively easy to automate or augment. As organizations deploy AI tools, they may choose to hire fewer junior employees while maintaining existing staff levels.

For commercial real estate, the implications extend beyond total employment counts. The more important question is how AI reshapes the composition of office-based work. If AI enables organizations to generate more output with fewer administrative employees while increasing demand for highly skilled technical workers, office utilization patterns could change significantly. Demand may become increasingly concentrated in locations that attract specialized talent, innovation ecosystems, and high-value knowledge work.

At this stage, AI's labor-market impact appears to be evolutionary rather than revolutionary. The technology is spreading rapidly, but measurable disruption remains limited. Much like the early days of ecommerce, the most significant effects may emerge gradually as adoption deepens, workflows evolve, and organizations learn how to integrate AI into core business operations. While concerns about displacement are understandable, current evidence suggests that AI is not yet driving broad-based unemployment. Instead, it is beginning to influence hiring patterns, workforce composition, and the future direction of employment growth.

The coming years will likely reveal whether these early signals develop into a more substantial labor-market transformation. For now, the data suggests that AI is reshaping the economy at the margins, with the greatest impact likely to come through changes in job mix rather than outright job destruction.

Sources & References

Anthropic. (2026). Labor market impacts of AI: A new measure and early evidence. https://www.anthropic.com/research/labor-market-impacts 

CBRE. (2026). AI’s Impact on the Economy, Employment & Productivity. https://www.cbre.com/insights/reports/ais-impact-on-the-economy-employment-and-productivity 

Goldman Sachs Asset Management. (2026). Weekly Market Monitor. https://am.gs.com/en-us/advisors/insights/article/market-monitor-weekly