🔍 Read the full analysis: What Keeps AI From Making The Workweek Shorter? on ThorstenMeyerAI.com
Get the latest gadgets delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
Thorsten Meyer’s November 18 report compares John Maynard Keynes’s 1930 vision of a 15-hour week with the persistence of the roughly 40-hour standard and early evidence on generative AI. It cites a reported 2.1% of work hours saved with generative AI in the United States in August 2026 and a Danish study finding no measurable effect on recorded hours; neither result shows that AI has shortened the workweek overall.
Generative AI is saving some workers time, but the evidence cited in a November 18 report does not show that it has shortened the workweek. Thorsten Meyer’s account places early AI findings against a longer history: US manufacturing hours fell sharply before 1930, then settled around a 40-hour standard, despite much higher incomes and generations of productivity growth.
The report says US workers saved about 2.1% of all work hours with generative AI in August 2026. It also cites a Danish study that found no measurable effect of AI chatbots on recorded hours. These findings describe different evidence: a time-saving estimate and an observed-hours result. Neither, on its own, establishes that workers have gained a shorter paid week.
Meyer uses economist John Maynard Keynes’s 1930 essay, “Economic Possibilities for our Grandchildren,” as a benchmark. Keynes predicted that living standards in what he called “progressive countries” would rise four to eight times over a century. Using the Maddison Project Database, the report calculates that real income per person by 2022 was about 4.4 times its 1930 level in Britain and 5.5 times in the United States.
Keynes’s frequently cited 15-hour week was not presented as a dated forecast. He suggested “three-hour shifts or a fifteen-hour week” as a possible way to share the remaining work, while predicting that people would still want some work. The report’s central distinction is that his income prediction is compared with historical data, while the shorter week was a proposal for distributing work and leisure.
Why AI Time Savings Aren’t Shorter Weeks
Whether AI reduces the hours people work is different from whether it helps them complete particular tasks faster. A reported 2.1% of work hours saved is a time-use estimate, not evidence that employers have cut schedules by that amount. The Danish finding, meanwhile, indicates that chatbot use had no measurable effect on recorded hours in that study; it does not establish that AI cannot change working time in other settings or over a longer period.
The distinction matters to workers because productivity gains can be used in several ways: employers may increase output, workers may take on other tasks, or the gains may eventually support reduced hours. The report documents the historical gap between economic growth and a shorter standard week, but it does not establish which choices explain that gap. Who receives the benefit of time saved remains a question of workplace practice and policy, not a result that follows automatically from new technology.
Products Worth Considering

Amazon Kindle Scribe Colorsoft 64GB with Premium Pen

Kindle Paperwhite Signature Edition 32GB (2024 Model) – Metallic Black

Amazon Kindle Scribe (16GB) with Premium Pen, Tungsten
As an affiliate, we earn on qualifying purchases.
From Factory Hours to the 40-Hour Standard
The historical series in Meyer’s report shows US manufacturing production workers averaging 51 hours a week in 1909, 49.4 in 1914, 45 in 1926 and 44.2 in 1929. The report links that decline to changes in working arrangements, including Ford’s move to an eight-hour day in 1914 and its announcement of a five-day, 40-hour week in 1926. These figures are specific to manufacturing workers, not all US employees.
Federal law later set a threshold that helped define the standard week. The Fair Labor Standards Act, signed in 1938, established a phased reduction in the maximum workweek before overtime pay applied: 44 hours in the first year, 42 in the second and 40 from October 1940. Work beyond the threshold remained possible, with overtime pay required. Meyer reports that manufacturing hours were 38.1 in 1940, 40.5 in 1950 and 39.8 in 1970; the October 2026 figure is 41.8 hours, described as preliminary.
The report also notes a contrasting European example: after a seven-week strike in 1984, Germany’s IG Metall union reached a settlement that reduced the metalworking week to 38.5 hours, followed by a gradual reduction to 35 hours in West German metalworking by 1995. This history shows that work hours can change through collective bargaining and law as well as productivity growth. It does not identify one cause for the persistence of the US standard.
What the AI Hours Data Can’t Show
The report does not provide enough methodological detail in the supplied material to assess how the 2.1% time-saving estimate was calculated, which workers or tasks it covers, or whether saved minutes translated into more output, less overtime or additional free time. The estimate is tied to August 2026; it should not be treated as a measure of a lasting change in work schedules.
The Danish study’s sample, measurement period and definition of recorded hours are also not specified here. Its finding of no measurable effect does not necessarily apply to every occupation or use of AI. The report’s October 2026 manufacturing-hours figure is marked preliminary and covers a sector-specific group of workers. It remains unclear whether AI-related savings will lead to shorter schedules, and the supplied material does not establish why the American workweek has stayed near 40 hours.
Evidence to Watch in Future Data
The next useful test will be whether repeated, comparable measurements show changes in actual scheduled and recorded hours, rather than only time saved on selected tasks. Readers can also watch for updated labor statistics and fuller details about the August AI estimate and the Danish study, including their samples, methods and time periods.
For now, the report offers a historical comparison, not a forecast that AI will—or will not—bring a shorter week. Any change would need to appear in workplace schedules, employment data or negotiated hours. Time saved on a task is not yet the same as time returned to workers.
Key Questions
Did Keynes predict that people would work 15 hours a week by 2030?
No. In his 1930 essay, Keynes offered a 15-hour week as a possible way to share the remaining work. Meyer’s report says it was not a dated prediction.
How much work time did generative AI save in the cited US estimate?
The report says US workers saved about 2.1% of all work hours in August 2026. The supplied material does not give enough methodological detail to determine how the estimate was calculated or whether it represents a lasting change.
Has AI shortened the workweek?
The evidence cited does not show that it has. The report gives a time-saving estimate and cites a Danish study with no measurable effect on recorded hours; neither result establishes a broad reduction in working hours.
What is the report’s figure for US manufacturing hours in October 2026?
It reports an average of 41.8 hours a week for production and nonsupervisory employees in manufacturing. The figure is identified as preliminary and does not describe all US workers.
Why did the report compare Keynes’s income prediction with the workweek?
The comparison highlights that living standards rose substantially in Britain and the United States while the US manufacturing week remained near 40 hours for decades. The report does not identify a single reason that higher productivity did not produce a shorter standard week.
Source: ThorstenMeyerAI.com
