AI Job Losses Now Measurable in US Data—Goldman Sachs Quantifies Real Impact

For months, economists debated whether AI would meaningfully disrupt labour markets or simply follow the pattern of previous technological shifts. Goldman Sachs has now provided the first rigorous quantification: AI is already a measurable drag on US employment, erasing roughly 16,000 net jobs per month over the past year.

Key Developments

Goldman’s analysis breaks down the dynamic clearly:

  • AI substitution effect: ~25,000 jobs eliminated monthly in roles easily automated
  • AI augmentation effect: ~9,000 jobs created monthly through productivity gains and new roles
  • Net impact: -16,000 jobs per month, translating to a 0.16 percentage point increase in the unemployment rate

The pain is concentrated. Entry-level workers and Gen Z—typically the cohort entering professional roles—face disproportionate displacement. This is the first time major financial research institutions have attached concrete numbers to AI’s labour market impact using actual employment data rather than models.

The tech sector is showing even sharper disruption: 52,050 job cuts in Q1 2026 represent a 40% jump year-on-year. In March alone, AI accounted for 15,341 layoffs—roughly 25% of total tech sector cuts, and a doubling of the AI-attributed share in a single month.

Industry Context: The Attribution Problem

While Goldman’s numbers are significant, Deutsche Bank economists have flagged an important caveat: “AI redundancy washing.” Companies are increasingly attributing layoffs to AI automation when the real drivers may be cyclical business pressures, restructuring, or simply cost-cutting optics.

This distinction matters for policy and workforce planning. If companies are hiding layoff reasons behind AI attribution, it obscures the true picture. If they’re not, we’re entering a period of genuine structural labour market disruption.

Practical Implications for Workers and Organizations

For entry-level professionals and recent graduates, the findings suggest:

  • Internship and junior roles are at highest risk of automation or elimination
  • Skill adjacency matters: roles that combine AI literacy with domain expertise are safer than pure-substitution roles
  • Sectoral exposure varies: knowledge-intensive sectors (finance, professional services, ICT) see faster displacement

For employers, the data validates investment in upskilling and augmentation strategies. Organisations adding 9,000 jobs monthly through AI augmentation are those pairing automation with human workers in complementary roles.

Open Questions

  • How durable is the augmentation effect? Will the 9,000 monthly jobs created through AI productivity gains persist, or does that number compress as AI systems mature?
  • Are we entering a structural shift or cyclical downturn? March 2026 AI layoff acceleration could signal either genuine automation or tech sector cost-correction.
  • What’s the lag between job elimination and creation? Workers displaced in substitution roles may lack skills for new augmentation roles—retraining pipelines haven’t scaled.
  • How will policy respond? No major labour market intervention has yet been announced at scale in the US, unlike some EU jurisdictions.

Source: Goldman Sachs Economic Research