🔍 Read the full analysis: Exploring Productivity Through The Lens Of AI on ThorstenMeyerAI.com
Get business pricing on tech for your team
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
ThorstenMeyerAI.com’s Nov. 12 article uses the history of productivity statistics to assess whether AI is raising output across the economy. It cites a study reporting a 15% productivity gain for support agents using AI, but says executives estimate AI has added about 0.2 percentage points a year to productivity growth so far.
ThorstenMeyerAI.com published an article on Nov. 12 examining whether artificial intelligence is lifting productivity across the economy, distinguishing reported gains within individual workplaces from a still-modest estimated contribution to overall growth. The article cites one study reporting a 15% productivity gain among support agents using AI, while saying executives put AI’s contribution to productivity growth so far at about 0.2 percentage points a year.
The article defines labor productivity as output per hour worked, a measure of how much value an hour of work produces. Its central argument is that this measure matters over decades because growth compounds: at 2% annual growth, output per hour doubles in about 35 years; at 1%, doubling takes about 70 years. These are illustrative growth-rate calculations, not forecasts of what AI will deliver.
To explain why new technology may take time to appear in broad statistics, the article revisits economist Robert Solow’s 1957 analysis of US output per worker-hour from 1909 to 1949. Solow attributed 87.5% of the increase to technical change and 12.5% to increased capital use, according to the figures quoted in the article. The article describes the technical-change component as including improvements beyond simply adding machines, such as better methods and organization.
The article says US productivity growth accelerated during the late 1990s, after computers had spread through workplaces. It reports an average annual increase of about 3.1% from 1995 to 2004, based on its calculation using a Bureau of Labor Statistics index. It also cites a 2000 estimate by Federal Reserve economists Stephen Oliner and Daniel Sichel that information technology use and computer production accounted for about two-thirds of the one-percentage-point acceleration they examined.
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’s Economy-Wide Gains Matter
The distinction between a task-level improvement and broad productivity growth matters for workers, employers and policymakers. If AI helps some employees complete more work in the same time, that can benefit particular organizations. Whether those gains spread widely, last, and show up in national statistics is a separate question. The article’s figures do not establish that AI has already changed economy-wide productivity in a lasting way.
Long-run productivity growth is closely connected to how much an economy can produce from each hour of work. If AI produces sustained improvements across many industries, it could contribute to higher output without relying only on longer working hours or more workers. But the source material’s own comparison argues against treating early workplace results as proof of a broad economic shift: the reported support-agent gain is a single study result, while the estimated economy-wide contribution remains small.
For readers, the practical takeaway is to watch both kinds of evidence. Company-level studies can show whether a tool helps with specific tasks; economy-wide measures can indicate whether those changes are accumulating across businesses and time. They answer related but different questions.
Products Worth Considering
Logitech MX Mechanical Wireless Illuminated Keyboard, Clicky Switches
Anker SOLIX C1000 Gen 2 Portable Power Station with 200W Solar Panel
Jackery Explorer 300 Portable Power Station with 40W Solar Panel, 292Wh
As an affiliate, we earn on qualifying purchases.
From the Computer Paradox to AI
The article places current AI claims alongside the history of computers. In a 1987 review, Solow observed that the computer age was visible in workplaces but not yet in productivity statistics. The remark became known as the Solow paradox. It captured a gap between the widespread presence of a technology and a measurable acceleration in output per hour.
That gap did not mean computers had no economic effect. The article reports that US productivity growth later rose to about 3.1% annually from 1995 to 2004, following slower growth in the preceding period. Oliner and Sichel’s estimate connected much of the acceleration they measured to information technology. The source does not establish that the later increase was caused by one factor alone; it presents the economists’ estimate as evidence about technology’s contribution.
The article also uses agriculture to illustrate how productivity can change the distribution of work. Citing US Department of Agriculture figures, it says farm output in 2023 was nearly three times its 1948 level, while total inputs had risen about 9% over that period. It links such increases in output per input to the long-term decline in farming’s share of employment. The comparison supplies historical context, not a prediction that AI will reshape jobs in the same way or on the same schedule.
“You can see the computer age everywhere but in the productivity statistics.”
— Robert Solow, quoted in his 1987 review of Manufacturing Matters
How Much AI Has Changed Output
The article does not provide enough methodological detail to evaluate the 15% support-agent result, including the study’s sample, duration, comparison group or definition of productivity. It also does not identify the survey or calculation behind executives’ estimate that AI has added about 0.2 percentage points a year to productivity growth. Those figures should be read as reported estimates, not as a settled measure of AI’s full economic impact.
It remains unclear how widely the reported workplace gains apply, whether they persist as organizations change their processes, and how much of any improvement will be recorded in official productivity data. The supplied article also ends partway through its discussion of the 2000 Federal Reserve analysis, leaving its full account of that estimate unavailable here.
Evidence to Watch in Productivity Data
The next test is whether reported gains spread beyond individual tasks and appear consistently across firms and industries. Readers can watch for new workplace studies that state their methods and comparison groups, alongside official output-per-hour data that track changes over time. No specific future release or timeline is identified in the source material.
For now, the article’s comparison with computers offers a caution about timing rather than a forecast: visible adoption and economy-wide productivity growth do not always rise together. The scale and durability of AI’s contribution remain open questions.
Key Questions
What does productivity mean in the article?
Labor productivity means output per hour worked: how much value is produced during an hour of work.
What AI productivity gain does the article report?
It cites one study reporting a 15% productivity gain for support agents using AI. The supplied material does not provide the study’s methods or enough detail to assess how broadly the result applies.
How much has AI added to overall productivity growth?
The article says executives put AI’s contribution at about 0.2 percentage points a year so far. It does not identify the survey or method behind that estimate, so it should not be treated as a definitive official measure.
What was the Solow paradox?
It refers to Robert Solow’s 1987 observation that computers were visible throughout workplaces but had not yet appeared clearly in productivity statistics. US productivity growth accelerated later, during 1995–2004, according to the figures cited in the article.
Does the article show that AI will cause a productivity boom?
No. It presents historical comparisons and early estimates, but does not establish that AI will produce a sustained, economy-wide acceleration. The scale, spread and duration of AI’s effects remain unclear.
Source: ThorstenMeyerAI.com
