Research

Work Analytics Lab

We study how work is performed, transformed, and augmented by technology, with a focus on task-level job analysis, AI impact, and labor market dynamics across organizations and geographies.

About Us

The MIT Work Analytics Lab (WAL) develops data-driven quantitative frameworks to understand how tasks, jobs, and organizations evolve in response to technological change, particularly artificial intelligence. By modeling work at a granular, task-based level, we analyze how AI interacts with human labor to shape productivity, job quality, and organizational performance. Our research spans private, public, and governmental organizations and combines large-scale labor market data, computational methods, and applied economic analysis. A central pillar of the lab’s mission is to provide actionable insights for firms, institutions, and policymakers while enabling rigorous comparisons across labor markets, especially between the United States and Europe.

Case Study Key Stats

%
Percentage of processes with AI acceleration potential
$
M
Estimated productivity gains from AI
$
Yearly productivity gains in year 1 of recommended AI use

Case Study: Workforce Planning in the Age of AI

This case study examines how the Company (ExampleCo) can use artificial intelligence to operate more efficiently amid rapid population growth, rising costs, and increasing demands on its transportation infrastructure. The study introduces a task-based methodology that breaks jobs into core activities to assess where AI can deliver the greatest value and how it will affect the workforce, providing data-driven guidance for strategic AI investment and workforce planning.
Read the Report

Research Areas

Organizational Dynamics and AI ROI Identification

We model organizations as dynamic systems to identify processes ripe for AI adoption with high ROI potential. Our frameworks assess productivity gains, worker career paths, and sustainability, helping organizations prioritize initiatives.

Job Profiles and Task-Level Work Modeling

We build detailed job profiles by decomposing roles into tasks, skills, and time allocation, enabling realistic representations of day-to-day work across industries and firms. This work supports applications in workforce planning, job redesign, and productivity analysis.

AI Exposure and Technology Impact Metrics

We develop novel metrics to measure exposure to AI and automation at the task, job, and organizational level. Our methods combine large-scale text analysis, expert evaluation, and technology signals to distinguish between augmentation, substitution, and task transformation. Our latest work examines the application of sentiment analysis to the design of exposure metrics.

Publications

AI and the Transportation Workforce

In this report, we aim to address executives, practitioners, and policy-makers in the field of transportation's most pressing question about AI and the workforce: How will AI impact the transportation workforce, what will the financial consequences be, and which parts of the workforce will be most affected? Read the report. 

Estimating the Task Content of Work: Workforce Design for AI-Driven Human–Robot Collaboration in Intralogistics

This paper addresses the challenge of strategic workforce planning for AI-driven human-robot collaboration (AI-HRC) in intralogistics. We ask two questions: how can task-level full-time equivalent (FTE) estimates be constructed from existing labor statistics, and how can these estimates, combined with AI exposure metrics, inform strategic AI-HRC design and workforce planning? Drawing on U.S. Bureau of Labor Statistics employment data, O*NET occupational profiles, and task-level AI exposure scores, we develop a stochastic task-time framework that decomposes occupations into tasks and models task frequencies as probability vectors on the simplex. Read the paper. 

Measuring the Intensive Margin of Work: Task Shares and Concentration

Are jobs diffuse bundles of activities, or are they concentrated on a small core? This paper develops a statistical framework to measure the intensive margin of work using task-frequency survey responses from O*NET. Our framework yields interpretable, budget-share-like task weights that aggregate transparently into standard economic outcomes, including exposure indices, and workforce or wage-bill decompositions. Empirically, we document a pronounced core-periphery structure of work: on average, the top three tasks account for about 31% of implied labor input across occupations. Read the paper. 

Across the Atlantic: Early Labor Market Responses Following the Introduction of AI in the United States and the European Union

We examine early labor market adjustments in the period associated with generative AI (Artificial Intelligence) introduction by comparing employment dynamics in the U.S. (United States) and the E.U. (European Union). Using large-scale workforce data linked to task-based AI exposure measures, we estimate within-firm employment reallocation by seniority and occupation while absorbing firm-level shocks. Across both regions, early-career employment contracts after 2022, with systematically larger relative declines in higher-exposure groups. Read the paper. 

News Sentiment as a Dynamic Predictor of Job Automation Risk

As artificial intelligence increasingly disrupts job and task structure, it is essential for companies and society, in general, to anticipate which tasks are at risk of automation and how these risks can guide workforce management strategies to proactively reskill employees, restructure roles, and optimize operations. To address these challenges, we introduce a machine learning pipeline that leverages news sentiment as a dynamic proxy for job automation risk assessment. By processing two million news articles, the model computes exposure scores at the task, job, and sector levels, enabling both historical trend analysis and real-time monitoring. Read the paper. 

People

See All People

Contact Us

Interested in collaborating with us? Reach out.
Email Us