TL;DR
AI has raised output in some individual workplace studies, including a reported 15% gain for customer support agents. But executives put AI’s effect on productivity growth across the economy at about 0.2% a year so far, and the source does not identify the survey or its methods. Past computer adoption also took time to appear in broad productivity figures.
A November 12 article from Thorsten Meyer AI says executives estimate artificial intelligence has added about 0.2% a year to productivity growth so far, even as one study it cites found a 15% productivity gain among support agents. The figures describe different kinds of evidence: results inside a particular workplace and an estimate of AI’s broader economic effect.
Get comfort and recovery gear delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
The article defines labor productivity as output per hour worked and argues that its growth shapes how quickly living standards can rise over time. It gives a comparison: at 2% annual growth, output per hour doubles in about 35 years; at 1%, it takes about 70 years. Those are illustrative calculations, not a forecast for any particular country or period.
The source points to a long-running measurement challenge. It says U.S. business output per hour in 2025 was about five times its 1948 level. Yet the adoption of computers did not immediately show up as faster overall productivity growth. The article says growth slowed after 1973, then rose to about 3% a year from 1995 to 2004; it attributes to Federal Reserve economists the estimate that information technology accounted for about two thirds of that acceleration.
For present-day AI, the article cites a study reporting a 15% gain for support agents, but it does not name the study or specify its setting in the supplied material. It also reports that executives put AI’s contribution to productivity growth at about 0.2% a year so far, without identifying the survey, sample, or calculation. The measures should not be treated as directly comparable: a gain for a defined group of workers is not itself an estimate of economy-wide growth.
AI’s productivity impact is still emerging
Workplace studies report gains, while executives estimate a modest effect on economy-wide productivity growth so far. These figures measure different things.
Two different measures
A result for one group of workers does not show how much the whole economy has changed.
- Reports a 15% productivity gain for support agents.
- The supplied material does not name the study or describe its methods.
- Executives put AI’s effect on productivity growth at about 0.2% a year so far.
- The survey, sample, and calculation are not identified.
Why productivity matters
Labor productivity means output per hour worked. Sustained growth can raise production without the same increase in hours.
This shows how much an economy produces for each hour of work.
The article’s illustrative calculations show that 2% annual growth doubles output per hour in about 35 years, compared with about 70 years at 1%.
Past technology offers context
Computer adoption did not immediately appear as faster overall productivity growth.
From Workplace Gains to Growth
Productivity matters because sustained increases in output per hour can raise the amount of goods and services an economy produces without requiring the same increase in hours worked. The article’s comparison between 1% and 2% annual growth illustrates how small differences compound over decades. It is a useful way to understand the stakes, though it does not establish that AI will produce either rate.
The gap between a reported workplace result and a small economy-wide estimate also shows why claims about AI’s economic impact need careful measurement. Tools may help a particular team complete tasks faster while adoption remains limited, or while firms take time to reorganize work around them. A single study cannot establish how much productivity has changed across industries or whether gains will persist.
That distinction matters to workers, businesses, and policymakers watching for evidence that AI is changing economic output. Broad productivity data can help show whether benefits extend beyond early adopters. The article presents the current estimate as modest and provisional, rather than evidence that AI has already transformed overall productivity.
Products Worth Considering
LEVOIT Vital 200S-P Smart Air Purifier for Large Rooms, White
Dyson Ball Animal 3 Upright Corded Vacuum
Bose SoundLink Max Portable Bluetooth Speaker, IP67 Waterproof, Black
As an affiliate, we earn on qualifying purchases.
Lessons From Earlier Technologies
The article uses economist Robert Solow’s research to explain why productivity is more than adding machines. In a 1957 paper, Solow examined U.S. output per man-hour from 1909 to 1949. The source quotes his finding that 87.5% of the increase was attributable to technical change and 12.5% to greater use of capital. That historical accounting separated measured capital deepening from a remaining contribution; it did not mean every part of that remainder had been individually identified.
It also describes changes in U.S. farming. According to figures attributed to the USDA’s Economic Research Service, farm output in 2023 was nearly three times its 1948 level, while combined inputs rose by about 9% over that period. The source gives average annual growth of 1.52% for output and 1.40% for total factor productivity. It links these gains with a fall in farming’s share of U.S. employment, from about one in eight workers in 1950 to an International Labour Organization estimate of 1.5% in 2025. The figures concern different measures and years, and do not by themselves explain every cause of employment change.
The article’s wider point is that technologies can take time to alter economy-wide statistics. It recalls that computers spread before a later productivity acceleration, which the source places between 1995 and 2004. This precedent offers context for AI measurement, but it cannot show whether AI will follow the same path.
“Gross output per man-hour “doubled over the interval, with 87½ per cent of the increase attributable to technical change and the remaining 12½ per cent to increased use of capital”.”
— Robert Solow, quoted in the article from his 1957 paper
The Scale of AI’s Effect
The source does not identify the study behind the reported 15% support-agent gain, including its sample, time period, or definition of productivity. It also gives no details about the executives behind the 0.2% annual estimate, such as how many were surveyed or how they assessed AI’s contribution. Readers cannot use the supplied information to judge the estimates’ methods or representativeness.
It remains unclear how quickly AI use will spread, whether early gains will hold across other tasks and industries, and how much of any measured change will reflect AI rather than other factors. The computer-era comparison is historical context, not proof that AI’s economy-wide effects will grow in the same way or on the same schedule.
Evidence to Watch Over Time
The article does not name a specific upcoming report or milestone. Assessing AI’s economic effect will require later productivity data and clearer information about how studies measure output, hours worked, and AI use. Readers can watch whether gains reported within particular firms or occupations are reflected in broader measures over time.
Further studies will also need to clarify which tasks and workers are covered, how long results last, and whether improved output per hour comes with changes in work quality or costs. Until those details are available, the estimates in the article describe an early and incomplete picture of AI’s productivity effect.
Key Questions
What does productivity mean in this article?
It means output per hour worked, also called labor productivity. It measures how much value is produced for each hour of work.
What productivity gain does the article report for AI?
It cites a study reporting a 15% gain for support agents, but the supplied source does not name the study or provide its methods. Separately, it says executives estimate AI’s effect on overall productivity growth at about 0.2% a year so far.
Does the 15% figure mean the whole economy is 15% more productive?
No. The source presents it as a result for support agents in one study. It is not an estimate of economy-wide productivity growth.
Why does the article compare AI with computers?
It says computer adoption preceded a period of faster productivity growth in the late 1990s and early 2000s. The comparison shows that broad effects can take time to appear in economic data; it does not establish that AI will follow the same pattern.
Source: Thorsten Meyer AI
