Hajra Naz
Massachusetts Institute of Technology released new research on AI workforce disruption. The study says artificial intelligence can already replace 11.7% of the U.S. labor market. That equals $1.2 trillion in wages across finance, health care, and professional services.
The work uses the Iceberg Index, a labor simulation tool built by MIT and Oak Ridge National Laboratory. The index models how 151 million U.S. workers interact across states, industries, and job roles. It tracks how AI, automation, and policy shifts influence tasks and skills in real time.
The Iceberg Index offers a forward-looking map of exposure. It highlights how AI will impact job markets far beyond coastal tech hubs. Lawmakers planning reskilling programs and AI workforce training can see disruption down to the ZIP code level.
“We are creating a digital twin for the U.S. labor market,” said Prasanna Balaprakash, ORNL director and research co-lead. ORNL runs the Frontier supercomputer, which powers large-scale modeling for the index.
The platform runs population-level experiments. It shows how AI reshapes skills, tasks, and labor flows long before those changes appear in the economy.
The system treats all 151 million workers as individual agents. Each agent carries data about skills, tasks, occupation, and location. It maps 32,000 skills, 923 occupations, and 3,000 counties, then checks which skills today’s AI can perform.
The research shows the visible tip of the iceberg. Layoffs and role shifts in tech represent only 2.2% of total wage exposure. Beneath that sits the full exposure of $1.2 trillion, tied to routine functions in HR, logistics, finance, and office administration. These areas often get ignored in automation forecasts.
The index is not a prediction tool. It gives a clear view of what current AI systems can already do. It helps policymakers test scenarios before spending billions.
State governments are already using the platform. Tennessee, North Carolina, and Utah validated the model with their own data. They also started building policy plans using the simulations.
Tennessee moved early. The state cited Iceberg in its official AI Workforce Action Plan. Utah leaders will release a similar report soon.
North Carolina State Sen. DeAndrea Salvador says the index exposes effects that traditional tools miss. She values the ability to drill down to local areas and match skills with automation risk.
The research challenges the belief that AI risk stays in big tech hubs. Iceberg simulations show exposure across all 50 states, including small towns and rural regions.
To close that gap, the Iceberg team built an interactive policy environment. States can test different moves, such as shifting training budgets or adjusting AI adoption strategies. They can measure how those choices affect GDP and employment.
The platform helps leaders pinpoint exposure hotspots, plan workforce investments, and test interventions before committing funds.
Balaprakash also shared state-specific findings with Tennessee’s governor and AI council. He said core industries like health care, nuclear energy, manufacturing, and transportation still rely on physical work, which slows full automation. The focus now is on using AI assistants and robotics to strengthen those sectors.
The team views Iceberg as an active simulation space. It gives states a structured way to prepare for AI’s impact on labor.
FAQs
1. What does MIT’s study say about AI and the U.S. workforce?
It says AI can already replace 11.7% of U.S. jobs, which is equivalent to $1.2 trillion in wages.
2. What is the Iceberg Index?
It is a labor simulation tool built by MIT and ORNL that models how AI affects skills, tasks, and occupations.
3. Which sectors face the highest AI exposure?
Finance, health care, logistics, HR, and office administration show strong exposure in the study.
4. Are only tech hubs at risk?
No. Iceberg simulations show exposure in every state, including rural regions.
5. How are states using the index?
States like Tennessee, Utah, and North Carolina use it to plan reskilling programs, test policy scenarios, and prepare for AI-driven shifts.
