Productivity and Artificial Intelligence

Artificial Intelligence has captured the interest of economists, policy makers, workers, and the public. How can we track the development of AI and its effect on workers and the economy? One metric to investigate is productivity. BLS productivity measures are useful for tracking changes in efficiency – the quantity of output produced for each unit of input. Changes in technology are one reason why productivity might increase over time. 

BLS produces productivity and related measures for the economy as a whole as well as for sectors and industries. This includes data on the inputs to production: labor, capital, energy, materials, and services.

Where to look in the data

BLS implicitly captures AI use through its capital measure of software used in production. Software is a part of the asset category Intellectual property products (IPP). 

As with any technological advancement there is a lot that goes into the production process. With AI, different industries will adopt the advancement in different stages. A lot of this process can be tracked using the industry data available on the BLS productivity tables page.

1. The total factor productivity by major sectors and total factor productivity by major industries tables show key variables including:
   1. Output growth
   2. Labor composition shifts in the labor force
   3. Capital input, growth in IPP
   4. Services input
   5. Energy Input
   6. Total factor productivity (TFP) – capturing effects of what can’t be measured (yet), such as joint influences of technological change, efficiency improvements, returns to scale, reallocation of resources, and other factors on economic growth.
2. While still seen in software related data, AI operations require Information processing equipment (IPE), the table includes measures for productive capital stock and capital investment and breaks down IPE into more detailed assets.
3. Intellectual property products (IPP) table includes measures for productive capital stock and capital investment and breaks down IPP into the detailed assets of software, research and development, and artistic originals.
4. The Rental prices table includes various measures of interest by detailed asset and industry including measures of asset productive capital stock, investment, and prices.
   1. For example, software is broken down into own-account, pre-packaged, and custom.
   2. Mainframe computers, integrated systems, communications equipment, storage devices, and industrial buildings.
   3. Note that variables in millions of constant dollars, like capital investment, productivity capital stock, and wealth stock, can be aggregated to create values for major industries and sectors.

AI and productivity publications and research

Exploring and measuring how AI is changing the economy ensures relevance in describing the dynamic U.S. economy and respond to evolving needs. 

Artificial intelligence (AI) is an emerging technological innovation in many aspects of production and across many industries. The U.S. Bureau of Labor Statistics (BLS) implicitly captures AI use through its capital measure of software used in production. A May 2026 article, "AI and the rise of software investment," uses data from the BLS productivity program to break down recent investment growth by asset category and, more specifically, to examine recent investment in the software category.

In addition, the following graphic shows how AI is moving through the economy, with each stop along the road associated with a chart below the graphic using productivity data. 

See the AI and productivity data presented in the AI Roadmap graphic and charts.

Other related research

• Integrated BEA-BLS Industry-Level Production Account, 1997–2024: The Sources of U.S. Economic Growth During the Expansion of AI (Corby Garner, Justin Harper, Matt Russell, and Jon Samuels). Survey of Current Business, April 27,2026.
• AI impacts in BLS employment projections. The Economics Daily, Bureau of Labor Statistics, March 11, 2025.
• Incorporating AI impacts in BLS employment projections: occupational case studies. (Christine Machovec, Michael Rieley, Emily Rolen). Monthly Labor Review, February 2025.

Last modified date: May 4, 2026