The headline figure โ 47 million U.S. jobs at significant risk of AI displacement โ comes from our analysis of 873 occupations using the AI Displacement Index. Here's exactly how that number breaks down and what it means for the American workforce.
The Headline Numbers
| Risk Tier | ADI Score | Jobs (millions) | % of Workforce | Median Wage |
|---|---|---|---|---|
| ๐ด Very High Risk | 81โ100 | 12.3M | 7.8% | $38,200 |
| ๐ High Risk | 61โ80 | 19.1M | 12.1% | $47,500 |
| ๐ก Elevated Risk | 41โ60 | 15.8M | 10.0% | $54,300 |
| ๐ข Moderate | 21โ40 | 62.4M | 39.5% | $61,800 |
| ๐ต Low Risk | 0โ20 | 48.3M | 30.6% | $52,100 |
The 47 million figure includes Very High + High + Elevated risk tiers โ workers in occupations where AI can automate 40% or more of core tasks within the next decade.
Top 20 Most Affected Occupations
| # | Occupation | ADI | Employment | Median Wage |
|---|---|---|---|---|
| 1 | Telemarketers | 97 | 127,000 | $29,600 |
| 2 | Tax Preparers | 95 | 63,800 | $46,300 |
| 3 | Data Entry Keyers | 94 | 152,600 | $36,200 |
| 4 | Bookkeeping Clerks | 91 | 1,512,000 | $45,500 |
| 5 | Loan Officers (Junior) | 89 | 328,000 | $63,400 |
| 6 | Customer Service Reps | 87 | 2,894,000 | $37,300 |
| 7 | Insurance Underwriters | 86 | 115,000 | $76,400 |
| 8 | Translators & Interpreters | 85 | 68,200 | $52,300 |
| 9 | Legal Secretaries | 84 | 172,000 | $48,800 |
| 10 | Proofreaders & Copy Markers | 83 | 11,200 | $42,100 |
| 11 | Payroll Clerks | 82 | 139,500 | $49,200 |
| 12 | Credit Analysts | 81 | 72,400 | $79,800 |
| 13 | Market Research Analysts (Junior) | 78 | 985,000 | $68,200 |
| 14 | Paralegals | 76 | 345,000 | $56,200 |
| 15 | Graphic Designers | 74 | 264,800 | $57,900 |
| 16 | Technical Writers | 73 | 55,200 | $79,900 |
| 17 | Travel Agents | 72 | 59,600 | $43,800 |
| 18 | Retail Salespersons | 71 | 4,080,000 | $30,600 |
| 19 | Cashiers | 70 | 3,338,000 | $27,300 |
| 20 | File Clerks | 69 | 58,500 | $35,700 |
By Industry Sector
| Sector | Jobs at Risk (โฅ41 ADI) | % of Sector | Largest Affected Role |
|---|---|---|---|
| Administrative & Support | 5.2M | 52% | Customer service representatives |
| Finance & Insurance | 3.8M | 48% | Bookkeeping clerks |
| Retail Trade | 6.1M | 39% | Retail salespersons |
| Information & Media | 1.4M | 45% | Editors & writers |
| Professional Services | 4.9M | 32% | Market research analysts |
| Healthcare | 3.2M | 16% | Medical records technicians |
| Manufacturing | 4.7M | 38% | Quality control inspectors |
| Transportation | 2.8M | 27% | Dispatchers |
By Pay Level
A critical finding: AI displacement is not limited to low-wage work. Unlike previous automation waves that primarily affected blue-collar jobs, AI reaches deep into middle-income and even upper-middle-income occupations.
| Income Bracket | Jobs at Risk | % of Bracket |
|---|---|---|
| Under $30K | 8.2M | 31% |
| $30Kโ$50K | 14.6M | 34% |
| $50Kโ$75K | 12.8M | 29% |
| $75Kโ$100K | 7.1M | 24% |
| Over $100K | 4.5M | 14% |
Timeline: When Does Displacement Hit?
- Already happening (2023โ2025): ~5.2 million jobs in data entry, basic content writing, customer service, and telemarketing are seeing measurable headcount declines.
- Near-term (2025โ2027): An additional ~12 million jobs in bookkeeping, junior legal, translation, and basic analysis roles face significant restructuring.
- Medium-term (2027โ2030): ~18 million more jobs across retail, mid-level finance, software testing, and graphic design face elevated risk as AI capabilities mature.
- Longer-term (2030โ2035): Remaining ~12 million jobs in complex professional services, healthcare administration, and advanced technical roles face increasing pressure.
What "At Risk" Actually Means
Important nuance: "at risk" does not mean "eliminated." Outcomes range across a spectrum:
- Full displacement (15โ25% of at-risk jobs): Role ceases to exist in current form
- Significant reduction (30โ40%): Companies need 50โ70% fewer people in the role
- Substantial restructuring (25โ35%): Role changes dramatically; some workers adapt, others don't
- Wage compression (10โ20%): Same headcount, but AI tools push wages down 15โ30%
Comparison with Other Estimates
| Source | Year | Estimate | Methodology |
|---|---|---|---|
| Frey & Osborne (Oxford) | 2013 | 47% of jobs | Expert surveys on occupation-level computerization |
| McKinsey Global Institute | 2023 | 30% of hours worked | Task-level analysis with GenAI capabilities |
| Goldman Sachs | 2023 | 300M jobs globally | Exposure analysis of generative AI |
| IMF | 2024 | 40% of global jobs exposed | AI exposure + complementarity index |
| AIJobWatch | 2025 | 47M U.S. jobs at significant risk | 5-pillar ADI methodology with real-time data |
The Bottom Line
47 million American workers โ nearly one in three โ hold jobs with significant AI displacement risk. This isn't a distant scenario; it's unfolding now. The question isn't whether disruption will happen, but whether we prepare for it in time.