A new study jointly released by the Massachusetts Institute of Technology (MIT) and Oak Ridge National Laboratory reveals that artificial intelligence technology is already capable of replacing 11.7% of the U.S. labor market—far exceeding the superficially observed 2.2%. Using the innovative “Iceberg Index” labor simulation tool, the research unveils the deep impact of AI on the job market.
Core Research Findings
According to CNBC’s report, this study analyzed data covering 151 million U.S. workers and found the potential impact of AI to be staggering:
- Directly Visible Impact: Current AI adoption is mainly concentrated in computing and technology fields, accounting for only 2.2% of the workforce, equivalent to approximately $211 billion in wage value
- Hidden Impact: When factoring in AI’s automation potential in administrative, financial, and professional services, the numbers jump to 11.7% of the workforce and about $1.2 trillion in wages
- Scale of Impact: 11.7% represents approximately 17.7 million workers, with wage value as high as $1.2 trillion
The Iceberg Index: Revealing Hidden Impact
The Iceberg Index developed by MIT and Oak Ridge National Laboratory simulates how 151 million U.S. workers interact across the country and how they are affected by AI and corresponding policies. The innovation of this tool lies in:
Two-Tier Analysis Framework
Surface Index
- Only observes current AI adoption
- Mainly concentrated in computing and technology industries
- Shows 2.2% workforce exposure
Iceberg Index
- Incorporates AI’s automation potential in cognitive work
- Includes administrative, financial analysis, and professional services
- Reveals the true impact scale of 11.7%
The research points out that if analysts only observe current AI adoption, they will severely underestimate its actual impact on the labor market. The Iceberg Index provides a more comprehensive assessment by considering both AI’s technical capabilities and economic feasibility.
Most Affected Industries and Positions
High-Risk Job Types
Fortune reports that jobs most susceptible to AI replacement are mainly concentrated in:
- Routine HR Work: Recruitment screening, payroll processing, employee record management
- Logistics Coordination: Supply chain management, inventory optimization, transportation scheduling
- Financial Analysis: Data analysis, risk assessment, financial reporting
- Office Administration: Document processing, schedule management, data entry
Industry Distribution Characteristics
The research found that AI impact is not evenly distributed across industries:
Financial Services
- Data analysis and risk assessment work highly automated
- AI can handle complex financial models and predictive analytics
- Customer service and financial consulting facing transformation
Healthcare
- Administrative and record management work impacted
- Diagnostic assistance systems replacing partial cognitive work
- Medical imaging analysis automation increasing
Professional Services
- Legal document review and contract analysis
- Accounting audit and tax preparation
- Consulting analysis and report generation
Geographic Distribution: Hidden Impact in the Rust Belt
The research revealed a surprising finding: Rust Belt states such as Ohio, Michigan, and Tennessee, while showing modest Surface Index values, demonstrate substantial Iceberg Index values.
Cognitive Work Supporting Manufacturing
The high Iceberg Index values in these traditional manufacturing states stem from cognitive work supporting manufacturing operations:
- Financial Analysis: Manufacturing financial planning and cost control
- Administrative Coordination: Production management and supply chain coordination
- Professional Services: Engineering support and quality management
According to Tech Startups report, Tennessee, North Carolina, and Utah have used their own labor data to validate the model and have begun building policy scenarios using the platform.
Technical Capability vs. Actual Impact
The MIT research team particularly emphasized that the 11.7% figure reflects technical capability and economic feasibility, not a prediction that these jobs will disappear on a set timetable.
Key Distinctions
Technical Feasibility
- AI currently possesses the technical capability to replace these jobs
- Algorithms and models can perform related tasks
- Accuracy and efficiency reach or exceed human levels
Actual Replacement Speed
- Enterprise AI adoption requires time and investment
- Organizational structures and workflows need adjustment
- Employee training and skill transition need support
- Regulatory policies may affect adoption speed
Fast Company analysis points out that the value of this research lies in helping policymakers and business leaders understand the true scale of AI impact, thereby formulating appropriate response strategies.
Industry Response and Policy Implications
Business Strategy Adjustments
Facing the labor market transformation brought by AI, businesses need to:
- Skills Retraining Programs: Invest in employee skill upgrades, cultivating AI-era capabilities
- Job Redesign: Redefine job roles, emphasizing human-machine collaboration
- Gradual Transformation: Avoid large-scale layoffs, adopt progressive AI integration strategies
- Ethical Considerations: Establish ethical frameworks and transparency standards for AI use
Policy Direction
The research provides important reference for policymakers:
Workforce Transition Support
- Establish comprehensive vocational retraining programs
- Provide career counseling and employment support services
- Invest in lifelong learning and skill development infrastructure
Social Safety Net Strengthening
- Assess adequacy of existing social security systems
- Explore new forms of work protection mechanisms
- Study innovative solutions like universal basic income
Regional Development Strategies
- Develop specialized policies for high-impact regions
- Support local economic diversification
- Invest in emerging industries and innovation ecosystems
Global Perspective and Taiwan Insights
While the MIT study focuses on the U.S. labor market, its findings have important reference value for global economies. For Taiwan:
Industry Structure Considerations
As a high-tech manufacturing powerhouse, Taiwan needs to pay particular attention to:
- Manufacturing Cognitive Work: AI impact on positions like engineering analysis, quality management, supply chain coordination
- Financial Services: Automation trends in banking, insurance, investment sectors
- Professional Services: Transformation needs in knowledge-intensive industries like law, accounting, consulting
Talent Development Strategy
Education and training systems need adjustment to cultivate AI-era capabilities:
- STEM Skills Enhancement: Strengthen science, technology, engineering, mathematics education
- Cross-disciplinary Capabilities: Cultivate critical thinking, creative problem-solving, interpersonal communication—capabilities AI cannot easily replace
- Lifelong Learning Culture: Establish social mechanisms for continuous learning and skill updating
Future Outlook
MIT’s Iceberg Index research provides an important warning: AI’s impact on the labor market is far more profound than it appears on the surface. The 11.7% workforce exposure and $1.2 trillion in wage risk highlight the importance of early planning and preparation.
As AI technology continues to advance, its scope of impact may expand further. Businesses, policymakers, and workers all need to actively respond to this transformation:
- Businesses should invest in employee training, redesign work processes, and achieve human-machine collaboration
- Governments should formulate forward-looking policies, establish social safety nets, and support workforce transition
- Workers should proactively learn new skills, adapt to change, and seize opportunities in the AI era
This is not just a technological change, but a deep transformation of the socioeconomic system. Only through multi-party collaboration can we ensure that the benefits of AI development are fairly shared while minimizing the impact.
