TL;DR
Get everyday essentials delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
A Nov. 10, 2026, essay by Thorsten Meyer AI examines how generative AI targets knowledge and clerical tasks, drawing on historical estimates that information work became a major part of the U.S. economy. The article says exposure to AI does not prove jobs will be eliminated; it reports that broad U.S. employment data shows no clear AI effect so far, while younger workers in highly exposed roles may be falling behind.
Thorsten Meyer AI published an essay Nov. 10 examining how generative AI is being applied to knowledge and clerical work, the large category of jobs built around information, communication and analysis. Drawing on historical economic estimates and recent labor figures, the essay argues that AI is reaching work that has become central to advanced economies, while stressing that exposure does not establish job loss.
The essay frames knowledge work as a distinct economic sector alongside agriculture, manufacturing and services. It points to economist Fritz Machlup’s 1962 study, which estimated that U.S. spending on producing and distributing knowledge reached $136.4 billion in 1958, or almost 29% of adjusted gross national product. Those figures describe spending, not the market value or quality of knowledge produced.
Machlup grouped activity into education, research and development, communications media, information machines and information services. The essay reports that education made up 44% of the total, media 28%, information services 13%, research and development 8%, and information machines 7%, using a breakdown tabulated by historian of statistics Benoît Godin. Machlup also estimated that workers in knowledge occupations increased from about 11% of the U.S. workforce in 1900 to 32% in 1959.
For a more recent comparison, the essay cites 2025 International Labour Organization figures showing managers, professionals and technicians accounted for 62% of employment in Singapore, 52.5% in the United States and 47.9% in Germany. It says higher-income and clerical roles are among the jobs most exposed to generative AI. It also reports no clear AI footprint in overall U.S. job data so far, alongside signs that younger workers in the most exposed jobs are lagging behind peers.
How AI Is Reshaping Knowledge Work and the Fourth Sector
Generative AI is targeting tasks across a large information economy. Exposure signals potential change to work, not proof that jobs will disappear.
What the fourth sector includes
Knowledge and information work spans industries that produce or distribute information.
Education
Machlup’s breakdown assigned education the largest share of knowledge-sector spending.
Research and development
Creating knowledge is part of the sector, alongside distributing it.
Media and information services
These activities share and deliver information to people and organizations.
Information machines
Tools and machines supporting information work were counted as part of the sector.
What the historical numbers measure
Machlup made knowledge work visible by counting related spending, not by measuring the value or quality of knowledge.
- Spending to produce and distribute knowledge in 1958
- Industries and occupations connected to information
- The value or effectiveness of knowledge produced
- How many jobs AI will eliminate or change
From AI exposure to workplace impact
AI may affect parts of a role while leaving other responsibilities in place.
Writing, reading, planning, calculating, checking and communicating may be affected.
Roles can still require judgment, accountability, interpersonal skills or domain knowledge.
Hiring, employment and wage data can help reveal sustained effects in exposed occupations.
AI Reaches the Information Economy
The development matters because AI tools are being aimed at tasks that make up a substantial share of work in wealthy economies: writing, reading, planning, calculating, checking and communicating. If software can perform parts of those tasks, its effects could reach beyond a narrow technology workforce and influence how organizations staff, train and manage many office roles.
But task exposure is not the same as a job being automated. A role may include work that AI can assist with while still requiring judgment, accountability, interpersonal skills or domain knowledge. The essay’s distinction is important for workers and employers weighing new tools: historical growth in knowledge work establishes the scale of the sector, but it does not predict how many jobs will disappear or how quickly responsibilities will change.
Popular right now
Vitamix VX1 Professional-Grade Blender, 64 oz, Black
De'Longhi Magnifica Duo Automatic Espresso Machine with Milk Frother – Silver
KitchenAid Artisan Mini 3.5-Qt Tilt-Head Stand Mixer, Candy Apple Red
As an affiliate, we earn on qualifying purchases.
How Economists Counted Knowledge Work
Management writer Peter Drucker used the term “knowledge worker” in his 1959 book The Landmarks of Tomorrow. As summarized in the essay, Drucker argued that workers who create value through learned expertise would require management practices different from those developed for factory production. His work provided a name for a shift that was becoming visible in offices, schools, research institutions and media organizations.
Machlup’s later effort to quantify that shift faced a measurement problem: knowledge cannot be counted in physical units like grain or steel. He instead calculated spending on industries and occupations connected to producing and distributing information. That approach made the sector visible in economic accounts, but it measured costs rather than the value or effectiveness of the knowledge produced. The essay uses this history to connect today’s AI debate with a longer change in the composition of work.
““Exposure is not job loss, yet.””
— Thorsten Meyer AI
Jobs, Tasks and AI Effects
The essay does not specify which U.S. employment dataset or time period supports its statement that overall job data shows no clear AI footprint. It also gives no figures or named study for the claim that younger workers in the most exposed roles are falling behind their peers. The size, duration and cause of that reported gap are therefore not established in the material provided.
It is also unclear how quickly AI adoption will change particular occupations, whether productivity gains will create new work, or how employers will distribute those gains. The historical figures describe past workforce and spending patterns; they do not measure the future effect of current AI systems.
Evidence to Watch in Hiring
The next test will be whether labor-market data shows sustained changes in hiring, employment or wages in occupations with high AI exposure. More detailed studies will be needed to separate AI’s effects from other economic forces and to identify differences by age, occupation and industry.
For employers and workers, the immediate question is how AI changes tasks within jobs, not simply whether a job title is labeled exposed. The essay offers no timeline or forecast for those changes. Future reporting will depend on clearer evidence about adoption, workplace outcomes and the experiences of workers in affected roles.
Key Questions
What is the “fourth sector” in the essay?
It refers to knowledge and information work, including occupations and industries that produce or distribute information, such as education, research, media and information services.What did Machlup’s 29% figure measure?
It was his estimate of spending on producing and distributing knowledge in the United States in 1958, expressed as almost 29% of adjusted gross national product. It measured costs, not the value of the knowledge produced.Does AI exposure mean a job will be eliminated?
No. The essay says exposure is not job loss. AI may affect some tasks in a role without removing the role itself.What evidence does the essay cite about current employment effects?
It says overall U.S. job data shows no clear AI footprint so far and reports that younger workers in the most exposed jobs are falling behind peers. The supplied material does not identify the data series, study or figures behind those statements.Source: Thorsten Meyer AI
