Quick Answer
So far, the end of work has been a myth: every wave of automation was predicted to end employment, and each time new work appeared to replace the old. But the prediction is not absurd, because AI targets cognitive work that earlier automation spared, and the real questions are how fast the transition happens, who bears the cost, and what work is for in the first place. The myth may become reality, or it may reveal that we never understood what work was for.
Key Takeaways
- ✦The end of work has been predicted since the Luddites and has never arrived.
- ✦New technologies destroy some jobs and create others, but the transition hurts.
- ✦AI differs from earlier automation by targeting cognitive and creative labor.
- ✦The deeper question is what work is for: income, meaning, identity, or all three.
- ✦Whether work ends is a political choice, not a technological necessity.
What Is the Claim?
The claim that work is ending is one of the oldest predictions in modern history, and it has never come true. The Luddites feared the machines would end their work; the economists of the 1930s feared technological unemployment; the automation prophets of the 1960s promised a leisure society by the year 2000. Each time, the work survived, usually in forms nobody predicted. The reason is a kind of economic gravity: when a machine replaces a task, the freed resources and lower costs create demand for new tasks, and the new tasks become the new jobs. The end of work keeps receding like a horizon.
The prediction persists anyway, because each new technology seems different. The computer, the robot, and now AI have moved up the ladder of human capability, and the current claim is that AI can do not just manual work but thinking work, and that this time the horizon is real. The question is whether the myth is finally becoming true, or whether it remains a myth because work is not the kind of thing that can end.
Historical Background
The phrase "end of work" was popularized by Jeremy Rifkin's 1995 book of that title, which predicted that automation would eliminate most jobs within decades. It joined a long line: Keynes's 1930 essay predicted a fifteen-hour week within a century, and the post-war futurists promised the leisure society. The record is consistent: the predictions were wrong in the aggregate, and employment kept growing, while the predictions were right in the particular, and specific occupations did disappear.
The economics of the pattern are now well understood. Automation replaces tasks, not whole economies; the cost savings create new demand; and the new tasks tend to be those that require judgment, care, and social interaction, things machines struggle to do. David Graeber added a darker observation: much of modern work is bullshit, jobs so pointless that even the workers do not believe they matter. The end-of-work debate is therefore also a debate about what work is for, and whether the work that would disappear would actually be missed.
The prediction has a politics, and the politics explains why it keeps returning. The promise of the end of work is liberating, the promise that the machines will do the toil and the humans will be free; and the threat of the end of work is frightening, the threat that the machines will take the jobs and the humans will be abandoned. The same forecast serves the optimist and the pessimist, and the forecasters on both sides are usually people who do not do the work they are predicting will end. The honest version of the prediction is the one that asks who is bearing the transition, who is being promised the freedom, and who is being told the jobs are gone.
Key Concepts
The first concept is the task-based view of automation. Occupations are bundles of tasks, and automation replaces the automatable tasks while raising the value of the rest. The doctor's diagnosis may be aided by AI, but the doctor's judgment, communication, and responsibility remain. The question is not which jobs survive but which tasks are left, and the answer has consistently been the tasks that require humans.
The second concept is the distinction between the end of jobs and the end of work. Jobs are the institutional form work takes under capitalism: employment, wages, contracts. Work in the deeper sense, purposeful activity, is older than jobs and will survive them. Even in a fully automated economy, people would do things: care, create, explore, build. The end of jobs is possible; the end of work, in this sense, is close to inconceivable.
The third concept is the meaning of work. Work provides income, but it also provides identity, structure, and social connection, and the end-of-work literature asks what would replace those goods. The utopians propose universal basic income and a society of leisure; the critics worry about purpose and dignity in a world without employment. The deeper question is whether work was ever really about the work, or about what work stands for: usefulness, belonging, and a place in the human community.
Contemporary Relevance
Generative AI has revived the prediction with new force. If a machine can write, code, design, and analyze, the argument goes, then the cognitive work that absorbed the last century of displaced labor is now being automated too, and the horizon may finally arrive. The evidence is mixed: adoption is real, but so is the creation of new kinds of work, and the bottlenecks are human judgment, regulation, and the sheer messiness of the world.
The practical conclusion is that the end of work, if it comes, will be a choice before it is a fact. Automation does not decide how the gains are distributed, whether the displaced are supported, or what the society values. The myth of the end of work is worth taking seriously, not because the prediction is likely true, but because the question behind it, what are we for if not for our work, is the most important question the automation era can ask.
The deeper question is the one the myth hides: what is work for? The economists count it as income; the sociologists count it as identity; the philosophers count it as one of the ways we make meaning. The automation that removes the drudgery is a liberation, and the automation that removes the meaning is a loss, and the same machine can do both, depending on what it is asked to do. The end of work, if it comes, will not be the end of purpose; it will be the demand that we find purpose somewhere else, in the play, the care, the craft, and the commitment that the work was always a stand-in for. The myth is worth keeping, not because it is true, but because the question behind it is the question of what we are for.
Sources
- Carl Benedikt Frey and Michael A. Osborne, "The Future of Employment" — https://doi.org/10.1016/j.techfore.2016.01.012
- David Graeber, Bullshit Jobs (Simon & Schuster) — https://www.simonandschuster.com/books/Bullshit-Jobs/David-Graeber/9781501143311
- John Danaher, Automation and Utopia (Harvard University Press) — https://www.hup.harvard.edu/catalog.php?isbn=9780674984106
Related Topics
- Future of Work — how labor is being reshaped.
- Philosophy of Economics — the theory of value and distribution.
- Work — the wisdom of work and its meaning.
- Automation vs Employment — the economics of the same question.
Continue Learning
Knowledge NetworkDeep Dive
Explore related concepts
- topic
The Future of Work
Related through Philosophy Of Economics
- wisdom
Work
Related through Philosophy Of Economics
- philosophy
Philosophy of Economics
Direct archive relation
- answer
Automation vs Employment: What Is the Relationship?
Related through Philosophy Of Economics
- answer
What is Technological Unemployment?
Related through Philosophy Of Economics
- answer
What is the Future of Work?
Related through Philosophy Of Economics
- answer
Will AI Replace Jobs?
Related through Philosophy Of Economics
- topic
Automation
Related through Work
Archive references
Sources
- 01The Future of Employment: How Susceptible Are Jobs to Computerisation?By Carl Benedikt Frey and Michael A. OsborneConsult source
- 02Bullshit Jobs: A TheoryBy David Graeber, Simon & SchusterConsult source
- 03Automation and UtopiaBy John Danaher, Harvard University PressConsult source
ZHAIBIAN Editorial Board reviewed
Reviewed by ZHAIBIAN AI Editorial Review · 2026-08-17