AIThis post was created with the assistance of artificial intelligence (AI).

TL;DR

A November 16, 2026, explainer from Thorsten Meyer AI examines how economists classify unemployment as AI changes work. It reports US unemployment at 4.1% in October, with 27.6% of unemployed people out of work for at least 27 weeks; those figures do not establish that AI caused the joblessness.

Thorsten Meyer AI published an explainer on November 16, 2026, outlining how economists distinguish frictional, structural and cyclical unemployment as debate grows over artificial intelligence’s effect on work. The article cites a 4.1% US unemployment rate in October 2026 and says more than one in four unemployed Americans had searched for work for at least six months, but it does not establish that AI caused those figures.

Buying for a business?Offer from Amazon

Get business pricing on tech for your team

  • Business-only prices and quantity discounts
  • Tax-exempt purchasing
  • Multiple users, one account, clear invoices
As an affiliate, we earn on qualifying purchases.

The article says the official US definition counts someone as unemployed if they had no job during the reference week, were available to start work and actively looked for a job during the previous four weeks. People who want work but stopped searching are excluded from the headline unemployment rate. The Bureau of Labor Statistics (BLS) supplies the definition and publishes the monthly survey results.

It describes frictional unemployment as short-term joblessness while people enter the labor market or move between jobs; structural unemployment as a longer-lasting mismatch between workers’ skills or locations and available jobs; and cyclical unemployment as job losses linked to economic downturns. In October, the median US unemployment spell was 11.9 weeks, while 27.6% of unemployed people had been searching for at least 27 weeks, according to the source article.

The article also reports that 36.0% of registered unemployed people in Germany—1,063,000 people—had been out of work for more than a year, citing the Federal Employment Agency. It cautions that German registered unemployment and the US survey use different definitions, so their levels should not be treated as directly comparable.

At a glance
analysisWhen: Published November 16, 2026; cites US l…
The developmentThorsten Meyer AI published an explainer on November 16, 2026, describing three types of unemployment and how to assess concerns about AI-driven job losses.
Labor markets · Briefing

Three Types of Unemployment As AI Changes Work

Unemployment can reflect job transitions, a mismatch between skills and available work, or an economic downturn. Recent figures show job-search duration, but do not establish that AI caused joblessness.

4.1%
US unemployment rate, October 2026
27.6%
US unemployed searching at least 27 weeks
11.9 weeks
Median US unemployment spell
36.0%
Registered unemployed in Germany out of work over a year

Three ways unemployment happens

The categories explain why one headline rate cannot show what is happening to workers.

Between jobs

Frictional

Short-term joblessness while people enter the workforce or move between jobs.

Skills or location

Structural

A longer-lasting mismatch between workers’ skills or locations and available jobs.

Economic downturn

Cyclical

Job losses linked to a downturn in the economy.

What the figures can tell us

Long searches matter to workers, employers and policymakers, but duration alone does not identify the cause.

US survey
  • Counts people with no job who were available and actively searched in the previous four weeks.
  • People who want work but stopped searching are excluded from the headline rate.
Germany registrations
  • Reports registered unemployed people, including 1,063,000 out of work for more than a year.
  • Different definitions mean the US and German levels are not directly comparable.

AI’s employment effect remains unmeasured

The article reports no estimate of AI-related job losses, roles created or changed, or AI’s contribution to long searches.

1
Track job searches

Watch whether long searches continue to rise, ease or differ across groups.

2
Follow worker transitions

Link changes in tasks and skills demand to hiring and workers’ job moves.

3
Separate evidence from cause

Unemployment levels and duration alone cannot show whether AI caused job losses.

Technology eliminates jobs, not work.— A US commission, as quoted in the article
The takeaway
Use the three unemployment categories to understand joblessness, and look for worker-level evidence before attributing it to AI.

How Joblessness Can Persist

The categories help explain why a single unemployment rate cannot show whether workers are between jobs, lack skills employers seek, or are losing jobs during a recession. That distinction matters to people weighing retraining, employers planning hiring, and policymakers considering how to respond to job losses.

The October figures cited by Meyer point to a sizeable share of long searches in both the US and Germany under their respective measures. They do not, by themselves, identify the cause. In particular, the article provides no evidence that AI accounts for the long-term unemployment share or the overall US rate.

For readers assessing AI’s impact, the distinction is practical: automation may remove tasks or change demand for particular skills, but whether those changes lead to sustained unemployment depends on what happens to workers and jobs across the wider economy. The article frames that question through established labor-market categories rather than treating every job change as technological unemployment.

Products Worth Considering

Bose QuietComfort Headphones (2nd Gen), Wireless Over-Ear Noise-Cancelling Headphones, Black

Bose QuietComfort Headphones (2nd Gen), Wireless Over-Ear Noise-Cancelling Headphones, Black

As an affiliate, we earn on qualifying purchases.

From Keynes to AI Concerns

The article opens with economist John Maynard Keynes’s 1930 warning about “technological unemployment,” the possibility that advances in technology could displace workers. It then recalls a 1966 US commission statement that “technology eliminates jobs, not work.” These historical references frame the recurring concern that automation can reduce particular jobs even as work changes.

It also cites research by David Autor and colleagues estimating that about 60% of US jobs in 2018 were in job titles that did not exist in 1940. That historical comparison indicates how job categories have changed over time; it does not prove that every displaced worker found a new role or predict how AI will affect employment.

The BLS says there was no monthly survey counting unemployed people before the 1930s. The modern survey began in 1940, and the BLS notes that its definition has changed little since. The source reports estimated unemployment of 24.9% in the US in 1933, based on historical statistics, and says the monthly survey is the basis for the figures now widely reported.

“Technology eliminates jobs, not work.”

— A US commission, as quoted in the Thorsten Meyer AI article

AI’s Role Remains Unmeasured

The figures in the article show unemployment levels and the duration of job searches; they do not isolate AI-related job losses. The source gives no estimate of how many people may have lost work because of AI, how many roles have been created or changed by it, or whether technology explains the rise in long-duration searches compared with 2019.

The US and German figures also come from different statistical systems: the US measure is based on an active-search survey definition, while the German number concerns registered unemployed people. The article warns against comparing their levels directly. It does not provide enough detail to establish which groups, industries or regions account for the reported long spells.

Watch Long-Term Job Searches

The article points readers toward unemployment duration as a measure to watch alongside the headline rate. Future data could show whether long searches continue to rise, ease or differ across groups, though those trends alone would not identify AI as the cause.

Assessing AI’s employment effects will require evidence that links workers’ job transitions to changes in tasks, hiring and demand for skills. The source offers no specific next release or forecast, so the scale and timing of any AI-related unemployment remain open questions.

Key Questions

What counts as unemployed in the US?

The BLS definition requires that a person had no job in the reference week, was available to start work and actively searched during the previous four weeks. People who want a job but have stopped looking are not included in the headline unemployment count.

What are the three types of unemployment?

Frictional unemployment occurs while people search between jobs or enter the workforce. Structural unemployment reflects a mismatch between workers’ skills or locations and available jobs. Cyclical unemployment is associated with job losses during economic downturns.

Does the October 2026 unemployment data show that AI is eliminating jobs?

No. The article reports a US unemployment rate of 4.1% and says 27.6% of unemployed people had searched for at least 27 weeks. It does not attribute either figure to AI or estimate AI-related job losses.

Can the US and German long-term unemployment figures be compared?

Not directly. The article reports that 27.6% of unemployed Americans had been searching for at least 27 weeks, while 36.0% of registered unemployed people in Germany had been out of work for more than a year. The measures use different definitions and time periods.

Source: Thorsten Meyer AI

You May Also Like

The New Worker Divide Is Between Plugged-In Setups and Patchwork Setups

Just as technology advances, the growing worker divide between seamless and patchwork setups may redefine productivity and opportunity—discover what’s coming next.

Public Perception of Automation: Fears and Realities

Keen insights into public fears and hopes surrounding automation reveal how society might face unforeseen challenges ahead.

Can Reskilling Keep Pace With Automation?

Great reskilling may bridge the automation gap, but understanding how to stay ahead requires exploring essential strategies and mindset shifts.