TL;DR
Get the little things that make your day delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
A November 10, 2026, episode of Thorsten Meyer AI’s “Work After Work” series examines knowledge work as a fourth economic sector and its exposure to generative AI. It draws on historical estimates and recent employment figures, while stressing that exposure to AI does not establish that jobs have been lost.
Thorsten Meyer AI published an episode on November 10, 2026, tracing the growth of knowledge work and examining why generative AI is aimed at many office tasks. The report uses historical estimates and recent labor-market figures to describe knowledge work as a major part of advanced economies, while cautioning that exposure to AI does not by itself show that jobs have disappeared.
The episode, the final installment in the first arc of the series “Work After Work,” frames knowledge work as a fourth sector alongside agriculture, industry and services. It revisits work that involves planning, reading, writing, calculating, checking and communicating—tasks that produce or pass along information rather than physical goods. The series argues that much of what is commonly grouped under “services” can be understood through this lens.
Its historical account centers on economist Fritz Machlup. In his 1962 book, The Production and Distribution of Knowledge in the United States, Machlup estimated that US spending on producing and distributing knowledge reached $136.4 billion in 1958, or “almost 29 per cent of adjusted GNP.” The figure reflects spending, not a direct measure of knowledge’s value. Education made up the largest listed share, at 44%, followed by media at 28%, information services at 13%, research and development at 8%, and information machines at 7%.
The episode also cites estimates of the growing workforce involved in information and professional work. It says Marc Porat found information workers’ share of the US workforce rose from under 6% in 1860 to 46% in 1970, while Machlup estimated that workers in knowledge occupations increased from about 11% in 1900 to 32% in 1959. For a contemporary comparison, it reports International Labour Organization figures for managers, professionals and technicians: 62% of jobs in Singapore, 52.5% in the United States and 47.9% in Germany. These categories are not necessarily identical to the historical definitions of knowledge work.
Exploring AI’s Role in Knowledge Work and the Fourth Sector
A November 10, 2026 episode examines how generative AI may affect information-based work—and why exposure to AI does not prove jobs have been lost.
What counts as knowledge work?
The fourth-sector framing groups information-focused tasks alongside agriculture, industry and services.
Create and share
Writing, reading and communicating produce or pass along information.
Plan and calculate
Planning, analysis and calculation are common parts of knowledge jobs.
Check information
Generative AI can be applied to summarizing and routine information handling.
Historical estimates, different definitions
These figures mark a long shift toward information work, but they are not one directly comparable series.
AI exposure is not job loss
The episode describes possible exposure in professional and clerical work, while leaving employment outcomes unresolved.
- AI may assist with tasks such as drafting, summarizing and analysis.
- Employers may change duties, hiring or team structures.
- That a role will disappear or how many jobs have been eliminated.
- That AI caused young workers in exposed jobs to fall behind peers.
What to watch next
Clearer labor-market evidence can help distinguish AI adoption from other forces shaping employment.
- Hiring and employment in occupations with greater AI exposure.
- Changes in job duties and pay as organizations adopt AI tools.
- Whether AI supports workers, replaces tasks or reshapes roles.
AI Meets the Office Workforce
The development matters because AI tools are being applied to tasks common in knowledge jobs, including drafting, summarizing, analysis and routine information handling. The episode says estimates place higher-income and clerical occupations among those with greater generative-AI exposure. That suggests the technology may affect work often associated with professional and administrative roles, not only manual or lower-paid jobs.
But the report draws a distinction between task exposure and job loss. A role may include activities AI can assist with without the position being eliminated. The source says overall US employment data has not shown a clear AI footprint so far, while also reporting that young workers in the most exposed jobs are falling behind peers. It provides no detailed figures or causal evidence for that reported pattern, so it should not be read as proof that AI caused the difference.
For workers and employers, the sector framing shifts attention from a simple count of jobs to the work performed within them. The effect of AI may depend on whether organizations use it to support employees, change job duties, reduce hiring or reorganize teams. The episode raises those stakes but does not establish which outcome will dominate.
Products Worth Considering
Philips Sonicare ProtectiveClean 5300 Rechargeable Electric Toothbrush with 3 Brush Heads, Black
Philips Sonicare 5950 Rechargeable Electric Toothbrush with 2 Brush Heads, Pink
Therabody SmartGoggles 2nd Gen Heated Eye and Temple Massager
As an affiliate, we earn on qualifying purchases.
How Knowledge Work Was Counted
The episode links the current AI debate to two earlier efforts to describe the rise of information-based work. Management writer Peter Drucker used the term “knowledge worker” in 1959, referring to people whose livelihoods depend on what they know. His argument, as presented by the source, was that managing people who create value through knowledge would require different practices from those developed for factory production.
Machlup’s 1962 study attempted to quantify knowledge production by counting its costs across industries and occupations. His categories included education, research and development, communication media, information machines and information services. The series notes a limitation: spending is not the same as value. The dollar total records resources devoted to knowledge production; it does not measure the quality or usefulness of the knowledge produced.
The source also points to a longer shift from factory-centered employment toward services and information work. Its estimates come from different studies, periods and occupational definitions, so the figures offer historical markers rather than one continuous, directly comparable measure.
“Exposure is not job loss, yet.”
— Thorsten Meyer AI episode
What AI Exposure Does Not Show
The figures presented do not establish how many jobs AI has eliminated, or how many positions may be created or changed. The source says there is no clear AI footprint in overall US job data so far, but does not specify the data series, time period or method used to reach that assessment.
The report also says young workers in the most AI-exposed jobs are falling behind their peers. It does not give the size of the gap, define the group or explain whether AI is responsible. The historical estimates are also based on differing definitions of information and knowledge work, making direct comparisons across decades uncertain. The episode does not provide enough evidence to determine how quickly AI adoption will affect particular occupations or which workers will bear the costs.
Evidence to Watch in Hiring
The immediate next step is to watch for clearer labor-market evidence: changes in hiring, job duties, pay and employment in occupations with greater exposure to generative AI. Those measures would help distinguish AI adoption from other forces shaping employment and show whether tools are augmenting work or replacing some tasks and roles.
The source identifies the episode as the final part of the first arc of “Work After Work” and says the following installment will examine what economic statistics measure—and what they miss. That discussion may help readers judge the limits of headline figures such as Machlup’s 29% estimate. No new employment forecast or policy measure is announced in the material provided.
Key Questions
What is the “fourth sector” in this report?
It is a way of describing knowledge and information work—activities such as planning, writing, calculating and communicating—alongside agriculture, industry and services. The source presents it as a framing of economic activity, not a universally adopted official sector classification.
What did Machlup’s 29% figure measure?
Machlup estimated that $136.4 billion in US spending in 1958 went toward producing and distributing knowledge, equal to almost 29% of adjusted GNP. The figure measures spending, not knowledge’s total value or quality.
Does AI exposure mean a job will be lost?
No. The episode distinguishes exposure to AI from job loss. AI may affect tasks within a role, but the material does not establish how many jobs have been eliminated or predict which occupations will shrink.
What does the source say about US employment so far?
It says overall US job data shows no clear AI footprint so far. It also reports that young workers in the most exposed jobs are falling behind peers, but supplies no detailed figures or evidence that AI caused that pattern.
Source: Thorsten Meyer AI
