🔍 Read the full analysis: Why Farming’s Past Offers Context For AI on ThorstenMeyerAI.com
Get business pricing on office and shipping supplies
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
A Nov. 3 article on ThorstenMeyerAI.com uses farming’s long decline in employment share to put AI-driven automation in historical perspective. It argues that agricultural change unfolded through successive technologies over generations, while leaving open whether AI will affect other sectors in a similar way or on a faster timeline.
Thorsten Meyer’s Nov. 3 essay on ThorstenMeyerAI.com uses the long decline of farm work to frame questions about AI and employment: U.S. agriculture’s share of workers fell from an estimated 73.7% in 1800 to about 1.5% today, but the shift unfolded over roughly two centuries rather than in a single wave. The history offers evidence that technology can sharply reduce a sector’s need for labor, but it does not establish how quickly AI will change work in other industries.
Meyer draws on historical estimates published by the U.S. Census Bureau. In Stanley Lebergott’s series, the U.S. labor force in 1800 included an estimated 1.4 million agricultural workers out of 1.9 million people aged 10 and older. The source emphasizes that figures before 1840 are estimates, not census counts, and that the series includes enslaved people. The 1810 estimate is higher, so Meyer describes the 1800 figure as roughly three workers in four rather than as a precise census result.
The article presents agricultural change as a sequence of developments: improved tools, including a steel plough made by John Deere in 1837; industrial production of ammonia fertilizer after Fritz Haber’s 1908 patent; and higher-yield crop varieties, including short-straw wheat associated with the Green Revolution in the 1960s. Citing Indian government statistics, Meyer says wheat output rose from 12.3 million tonnes in 1964–65 to 26.4 million tonnes in 1971–72, then to a third advance estimate of 117.5 million tonnes for 2024–25.
The decline in agriculture’s share of employment did not mean an immediate fall in the number of farm workers. Meyer says the U.S. farm workforce continued to grow until about 1910, when it reached 11.8 million. He also cites a German historical series in which agriculture’s employment share fell from 43% in 1882 to 25% in 1939 while the number of farm workers remained between 8 million and 10 million. The essay’s global figures, attributed to a 2025 International Labour Organization estimate, put the number of people working in agriculture at 928 million, or about one in four workers worldwide.
Why Farm Automation Took Generations
The history complicates simple predictions that automation will either eliminate work quickly or leave employment unchanged. Farming’s labor share fell as tools, fertilizers and crop varieties raised output, but the process took generations and the number of workers initially continued to rise. That distinction matters for people judging claims about AI: a falling share of workers in a field does not necessarily mean an immediate fall in the headcount, and technological change can alter productivity before it substantially changes employment.
Meyer uses food production as a case study in how economies can meet demand with far fewer workers in one sector. He cites a 4.2-fold increase in global cereal harvest since 1961, compared with a 2.7-fold increase in population, and says U.S. farm output in 2023 was nearly three times its 1948 level with only a small rise in total inputs. These comparisons support the essay’s point about productivity, but they do not prove that AI will produce the same employment pattern or output gains in offices, factories or other services.
The essay’s relevance is therefore less a forecast than a historical caution. Automation can change what workers do and where labor is needed, yet the transition depends on technology, markets and the availability of other jobs. Meyer’s comparison gives readers a longer timescale for considering AI’s possible effects without treating farming’s history as a direct prediction.
Products Worth Considering
Apple MacBook Air 13-inch with M5 chip, 16GB memory, 1TB SSD, and 12MP Center Stage camera (Midnight)
Samsung 34-Inch Odyssey G5 Ultra-Wide Curved Gaming Monitor
Canon PIXMA TR160 Wireless Portable Color Inkjet Printer
As an affiliate, we earn on qualifying purchases.
From Muscle Power to Higher Yields
Meyer’s article is the second installment of “Work After Work,” a series examining how economies are built. Its first episode divided work into field, factory and office; this installment focuses on farming because, in Meyer’s framing, it is the clearest example of a sector whose employment share has fallen dramatically.
The article says early farming depended heavily on human and animal labor, limiting how much food a household could produce beyond its own needs. It notes that historical estimates for other countries also show large agricultural or primary-sector workforces: about 85% in Japan in 1872, 83.5% in China in 1952, and 69.7% of workers in India in 1951. These figures come from different historical reconstructions and national statistical series, so their methods and coverage are not identical.
Several specific milestones illustrate the gradual process. Deere & Company’s account dates the steel plough to 1837. The Nobel Foundation links the spread of dwarf wheat varieties to the work of Norman Borlaug and other plant breeders in India and Pakistan in the mid-1960s. Meyer also cites a 1908 patent by German chemist Fritz Haber for synthesizing ammonia, followed by industrial production. Together, the examples show a mix of mechanical, chemical and biological change—not a single invention that suddenly removed farm labor.
“Farming shows a whole sector can be emptied of human hands. It also shows that it took about 200 years in America.”
— Thorsten Meyer, ThorstenMeyerAI.com
What History Cannot Predict About AI
Meyer’s suggestion that automation could affect as much as 90% of a sector within a decade or a few decades is a personal thesis in the essay, not a measured forecast or a conclusion established by the farming data. The article does not provide a method for estimating that figure, identify which sector it would apply to, or specify how many jobs AI might replace, change or create.
The historical figures also carry limits. Some are estimates or reconstructions, and national definitions and counting methods vary. The essay’s long-run output comparisons show what happened to agricultural production, but do not isolate the contribution of any single technology or establish that comparable gains are likely in other industries. It remains unclear how AI’s adoption, labor effects and timeline will compare with the gradual changes in farming.
The Series Turns to Factory Work
Meyer says the next episode of “Work After Work” will examine mechanization in factories, including reapers, threshers and tractors as part of the broader story of agricultural change. That installment may add context on how machinery altered farm labor beyond the tools and crop innovations discussed here.
For now, the essay’s evidence supports a bounded conclusion: farming’s employment share fell dramatically as productivity rose, but the change took a long time and did not follow a simple, immediate path. Readers looking for a clearer account of AI’s effects will still need evidence on adoption, employment and productivity in the specific industries affected.
Key Questions
What is the main point of Meyer’s article?
Meyer uses agriculture’s long shift toward mechanized, higher-output production to show that automation can reduce a sector’s share of employment, while stressing that the historical change took generations and is not a direct prediction for AI.
How many Americans worked in agriculture around 1800?
The historical series cited in the essay estimates that 73.7% of the U.S. labor force worked in agriculture in 1800. The figure is an estimate, not a census count, and the source says it includes enslaved people.
How many people still work in agriculture worldwide?
Meyer cites an International Labour Organization estimate of 928 million people in 2025, about one in four workers worldwide. This is a global estimate, not a count of people employed only in crop farming.
Does farming’s history show that AI will eliminate jobs?
No. The article offers a historical comparison, not a forecast. It describes a long decline in agriculture’s employment share and does not establish how AI will affect jobs, wages or productivity in other sectors.
What does Meyer mean by the 90% automation figure?
It is Meyer’s stated thesis that automation could reach a sector of up to 90% within a decade or a few decades. The article does not present the figure as a confirmed statistic or provide a detailed forecast method.
Source: ThorstenMeyerAI.com
