🔍 Read the full analysis: Rethinking Knowledge Work With AI on ThorstenMeyerAI.com
Get smart everyday buys 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
ThorstenMeyerAI.com’s November 10 article examines the growth of knowledge work and its exposure to generative AI. It cites historical estimates that information and knowledge work had become a major part of the US economy, while stressing that exposure does not prove job loss and that broad US employment data has not yet shown a clear AI effect.
ThorstenMeyerAI.com published an installment on November 10 examining how knowledge work grew into a major part of the economy and why generative AI is aimed at many office tasks. The article draws on economic studies by Fritz Machlup and Marc Porat, while cautioning that exposure to AI is not the same as job loss and that US employment data has not yet shown a clear overall AI footprint.
The installment, Episode 8 of the series “Work After Work,” starts with a question about how much of a working day is spent on tasks such as meetings, planning, reading, writing, calculating and checking. It connects those activities to what economist Fritz Machlup called the production and distribution of knowledge. Machlup estimated that US spending on knowledge-related activities reached $136.4 billion in 1958, or “almost 29 per cent of adjusted GNP.” That figure measures spending, not the market value or productivity of knowledge itself.
The article also draws on Marc Porat’s 1977 study for the US Department of Commerce, which classified workers by what they did rather than by the industry where they worked. Porat’s estimates put information workers at 5.8% of the US workforce in 1860 and 46.4% in 1970. The source says information work had become the largest worker group by the mid-1950s. These are historical classifications and estimates, not a current count of AI-exposed workers.
For a present-day comparison, the article cites 2025 International Labour Organization figures showing managers, professionals and technicians accounted for 62% of jobs in Singapore, 52.5% in the United States and 47.9% in Germany, amid the changing nature of work. It says higher-income and clerical roles are among those most exposed to generative AI, but exposure alone does not establish that jobs will disappear. It also reports that overall US job data has not shown a clear AI effect so far, while younger workers in the most exposed jobs are falling behind their peers.
Infographic 01 · The number to remember
Fritz Machlup's first count of the knowledge industries, by branch.
US spending on knowledge by branch, 1958 · share of $136.4 bn
Machlup measured what knowledge cost (spending, the way national accounts count), not what it was worth. “Almost 29 per cent of adjusted GNP” is his figure for 1958. The branch values are as tabulated by Godin (2008); the labour shares are Machlup’s estimate as summarised by his publisher.
Sources: F. Machlup, The Production and Distribution of Knowledge in the United States (Princeton University Press, 1962), as tabulated by B. Godin, “The Knowledge Economy: Fritz Machlup's Construction of a Synthetic Concept” (2008); Princeton University Press, book description.
Powered by Thorsten Meyer AI
AI Reaches the Office Economy
The article’s central point is that AI is directed at a large category of work that has often been less visible than factory or agricultural labor. Tasks such as drafting, reading, calculation and information handling are part of many professional and clerical roles, so changes to them could affect a substantial share of workers. The cited figures indicate the scale of knowledge and information work, though they do not measure how many tasks current AI systems can perform reliably.
For readers, the distinction between exposure and job loss is consequential. A tool may change how work is done without eliminating a role; employment effects may vary by occupation, employer and the way organizations adopt the technology. The source’s note about younger workers in more exposed jobs points to a possible early signal, but it does not establish that AI caused the gap or quantify its size.
The historical framing also raises a question about how organizations manage work built around expertise and judgment. The source attributes to Peter Drucker the argument that managing people who create value through knowledge calls for different rules from those developed for factory production. That is relevant as companies consider whether AI will supplement workers, reorganize responsibilities or reduce hiring.
Products Worth Considering
Logitech MX Master 4 for Mac Wireless Ergonomic Mouse, White Silver
Logitech MX Master 4 Ergonomic Wireless Mouse with Haptic Feedback, Graphite
Samsung 34-inch ViewFinity S50GC Ultra-WQHD Monitor
As an affiliate, we earn on qualifying purchases.
From Knowledge Work to Information Jobs
Peter Drucker used the term “knowledge worker” in 1959, according to the source, to describe people who earn a living through what they have learned rather than primarily through manual labor. His argument, as presented in the installment, was that the growth of mind-based work would require management practices beyond the factory systems associated with assembly-line production.
In 1962, Machlup attempted to measure the economy involved in producing and distributing knowledge. He grouped activity into education, research and development, communications media, information machines and information services. The source says education represented 44% of his estimated 1958 total, while media accounted for 28%. These shares describe Machlup’s accounting categories and spending estimate, not the proportion of modern jobs or output attributable to AI.
Porat later applied an occupational lens, separating the economy into agriculture, industry, services and information. His approach counted information workers even when they worked in other industries—for example, a clerk in a steel mill. Together, these accounts explain the source’s idea of a fourth, or quaternary, sector: work centered on information and knowledge.
““Exposure is not job loss, yet.””
— ThorstenMeyerAI.com, summarizing the distinction between AI exposure and job loss
AI’s Employment Effects Remain Unclear
The source does not give a study name, data window or measurement method for its statement that US job data shows no clear AI footprint. It also does not specify the size of the reported gap between younger workers in the most exposed jobs and their peers, or establish that AI caused it. Those points should be treated as claims in the source article rather than a quantified demonstration of AI’s labor-market effects.
It is also unclear which specific occupations, tasks or groups of workers the article counts as most exposed, and what level of AI capability or workplace adoption underlies that assessment. The historical estimates from Machlup and Porat use different definitions and periods; they cannot be read as direct measures of today’s AI exposure. The source does not provide a forecast for net job gains or losses.
The Series Turns to Economic Measures
The source identifies this installment as the final episode of the series’ first arc. It previews a subsequent episode about what economic statistics measure and what they leave out. That topic follows from Machlup’s approach: his 29% estimate counted the costs of knowledge-related activity rather than directly measuring knowledge’s value.
For the AI question, the next evidence to watch is clearer, occupation-level data tracking adoption, task changes, hiring and employment over time. The article does not announce a specific release date or provide a timetable for such data, so the timing and scale of any measurable labor-market effects remain open.
Key Questions
What is the news in this article?
ThorstenMeyerAI.com published an installment on November 10, 2026, tracing the rise of knowledge work and discussing why generative AI is aimed at office and information tasks.
What did Machlup estimate?
Fritz Machlup estimated that US spending on producing and distributing knowledge was $136.4 billion in 1958, or almost 29% of adjusted GNP. The source notes that this was a spending measure, not a direct measure of knowledge’s value.
Does AI exposure mean a job will be eliminated?
No. The source explicitly distinguishes exposure from job loss. It does not provide evidence that every exposed task or occupation will disappear.
What does the source say about current US employment data?
It says overall US job data has not shown a clear AI footprint so far, while younger workers in the most exposed jobs are falling behind peers. The article does not specify the data window or establish that AI caused the reported difference.
What comes next in the series?
The source says the next episode will examine what economic numbers measure and what they miss. It does not give a publication date in the supplied material.
Source: ThorstenMeyerAI.com
