🔍 Read the full analysis: Where AI Meets Productivity: A Guide For Work on ThorstenMeyerAI.com
Get business pricing on office and shipping supplies
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
A November 12, 2026, article from ThorstenMeyerAI.com examines whether AI will raise productivity, measured as output per hour worked. It cites gains in individual workplace studies but says executives estimate AI has added about 0.2 percentage points a year to productivity growth so far; how much those gains will spread across the economy remains uncertain.
ThorstenMeyerAI.com published a guide to AI and workplace productivity on November 12, comparing reported gains in individual firms with the much smaller estimated effect on overall productivity growth. The article says executives put AI’s contribution to productivity growth so far at about 0.2 percentage points a year, while one study it cites found a 15% productivity increase among customer-support agents. The gap highlights how local improvements may take time to register across the economy.
The installment, part of the site’s “Work After Work” series, defines labor productivity as the value produced per hour worked. It argues that sustained productivity growth matters because it compounds: the article estimates output per hour doubles in about 35 years at 2% annual growth, compared with about 70 years at 1%. These are illustrative calculations, not forecasts of household income.
The piece uses historical US data to explain why new technology does not automatically cause an immediate economy-wide jump. It cites economist Robert Solow’s 1957 analysis of output per worker-hour from 1909 to 1949, which attributed 87.5% of the increase to technical change and 12.5% to increased capital use. The article also reports that US nonfarm business productivity grew about 1.4% annually from 1973 to 1995, then about 3.1% annually from 1995 to 2004, based on its calculations from a Bureau of Labor Statistics index.
For the current AI period, the article contrasts the support-agent study’s 15% workplace-level gain with executives’ estimate of AI’s contribution to aggregate productivity growth. It does not provide, in the supplied material, the study’s full methodology or the survey details behind the 0.2-point estimate. The figures describe different measures and should not be read as directly comparable.
Infographic 01 · The sector clock
Productivity moves the sector clock
US farms produced nearly three times their 1948 output in 2023, with only about 9% more inputs in total. Over the same decades, farming's share of US jobs fell from about one in eight to 1.5%.
- Field12.1%
- Factory29.9%
- Services58.0%
Lebergott · own calculation
- Field1.5%
- Factory18.9%
- Services79.6%
ILO estimate
Farm output +210% (1948-2023, 1.52% a year); all inputs +0.12% a year (about +9% in total); total factor productivity +1.40% a year (USDA ERS). 1950: services = remainder, incl. unemployed.
Sources: USDA ERS, Agricultural Productivity in the U.S. (updated 25 Aug 2026); US employment shares: S. Lebergott, in US Census Bureau, Historical Statistics of the United States (1975; 1950, own calculation, see episode 2); ILO modelled estimates via World Bank (2025).
Powered by Thorsten Meyer AI
Why AI Gains May Take Time to Spread
The distinction between productivity at one firm and productivity across an economy matters to workers, businesses and policymakers. A tool can help a particular team complete tasks faster without changing national output per hour much: adoption may be limited, the time saved may go to other tasks, or firms may need to reorganize work before improvements show up in measured output.
The historical record in the article cautions against treating early enthusiasm as proof of a lasting productivity boom. Computers were widespread before US productivity growth accelerated in the late 1990s. Conversely, the later improvement does not establish that AI will follow the same path. The central issue is whether localized AI gains become broad, sustained output growth, not simply whether workers use the tools.
That distinction also affects how the public should interpret AI claims. A reported percentage gain in a controlled or firm-specific study can be informative, but it does not by itself show how much an entire economy produces per hour. The source’s figures suggest early effects worth tracking, while leaving the eventual scale and timing unsettled.
Products Worth Considering
Logitech MX Master 4 for Mac Wireless Ergonomic Mouse, White Silver
Samsung 32 Odyssey G5 (G51F) QHD 180Hz Gaming Monitor
Vari Ergo Electric Standing Desk, 54×26, Light Wood/White
As an affiliate, we earn on qualifying purchases.
From the Solow Paradox to AI
The article draws on the productivity paradox associated with Solow. In a 1987 New York Times Book Review passage quoted by the source, Solow observed that a perceived technological revolution had coincided with slower productivity growth. His remark that computers could be seen everywhere except in productivity statistics became a shorthand for the delay between adopting technology and seeing its effects in economic data.
The source then points to the US productivity acceleration from 1995 to 2004. It cites a 2000 estimate by Federal Reserve economists Stephen Oliner and Daniel Sichel, who attributed about two-thirds of a one-percentage-point increase in productivity growth to information technology use and computer production. That historical finding offers a comparison, not a template that proves AI will produce a similar result.
The article also uses agriculture to show how productivity can change the distribution of work. It says US farm output in 2023 was nearly three times its 1948 level, while total inputs had risen by about 9% over the period, citing USDA data. The example illustrates that higher output can result from improved methods and technology rather than simply adding workers or equipment.
“Productivity isn’t everything, but in the long run it is almost everything.”
— Paul Krugman, as quoted by ThorstenMeyerAI.com
What the Early AI Figures Cannot Show
The figures presented do not establish AI’s long-term effect on economy-wide productivity. The source gives an estimate of about 0.2 percentage points a year based on executives’ views, but the supplied material does not identify the survey, its sample, or the period covered. It also cites a 15% gain for support agents without providing the study’s title, design, or broader applicability.
It remains unclear how widely such gains have been adopted, whether they persist outside the studied workplaces, and how productivity statistics will capture changes in work quality or task mix. The article’s historical comparison with computers shows that technology’s measured effects can be delayed, but it cannot determine whether AI will produce the same pattern or how large any eventual increase might be.
Tracking Adoption and Output per Hour
The next test is whether workplace-level gains spread across businesses and appear in productivity data over time. That requires tracking both adoption and measured output per hour, alongside changes in work processes and investment. The source does not identify a specific upcoming report or release date for an updated AI productivity estimate.
Readers should also watch how future studies define productivity, which workers and industries they cover, and whether their comparisons use a clear baseline and time window. Until comparable evidence accumulates, the 15% study result and the 0.2-point estimate should remain distinct indicators—not proof of a broad productivity surge.
Key Questions
What does productivity mean in this article?
It means output per hour worked: how much value workers produce in an hour. Economists also call this labor productivity.
How much has AI increased productivity so far?
The source says executives estimate AI’s contribution to productivity growth at about 0.2 percentage points a year so far. It also cites one study reporting a 15% gain for customer-support agents. These figures measure different things and do not establish a single economy-wide effect.
Does the support-agent result mean all workers are 15% more productive?
No. The source describes a gain in one study involving support agents. Without details about the study’s methods and scope, the result should not be generalized to all workers or industries.
Why does the article compare AI with computers?
It uses the history of computers to show that widespread technology adoption may precede a measurable rise in economy-wide productivity. The comparison provides historical context; it does not show that AI will follow the same path.
Source: ThorstenMeyerAI.com
