Will Robots Take Over Our Jobs?

The Forecasts, and What They Actually Say

Free write (5 min): will robots take your job? Students have absorbed a great deal about this without examining any of it, so ask where the impression came from as well.

The vocabulary, which students use loosely and the deck defines precisely: automation (technology that minimises human input), artificial intelligence (machines programmed to simulate human thinking), machine learning (programs that learn from new data without human assistance), upskilling (expanding a worker's skills in their current role) and reskilling (equipping a worker for a substantially different role).

The headline forecasts:

• An Oxford study of 702 jobs found 47% could be done by machines within a decade or two.
• An OECD working paper found 46% of jobs across 32 countries have a better-than-even chance of automation — 210 million jobs.
• The World Economic Forum projected that work done by machines would rise from 29% to over 50% by 2025.

Then the slide that reframes all three, and it is the most important one in the deck. A McKinsey study found that about half of all workplace activities could be automated with existing technology — but fewer than 5% of occupations are candidates for full automation, while almost every occupation has significant partial automation potential.

Assign the readings, due before Tuesday:

• Why 'the Future of AI Is the Future of Work' — MIT Sloan
• The Rise of the Machines — Why Automation Is Different This Time — In a Nutshell (video)
• Don't Fear AI. It Will Lead to Long-Term Job Growth — World Economic Forum
• Will Automation Take Away All Our Jobs? — David Autor (video)

Read the Charts Carefully

Mindful moment (5 min), then four charts that each require a careful read:

• Automation risk against GDP per person. The vertical axis is the share of jobs at risk of full automation, where "at risk" means a 50% chance or higher. The relationship with national income is negative — richer countries face less risk. Ask why two countries with similar incomes might still differ: the answer is industry mix.
• Which industries. Food preparation is most at risk, then construction, cleaning, and driving. The pattern students usually spot is manual work — but push them, because personal service, sales, and customer service are service jobs at moderate risk too.
• Which activities. This is what drives the industry numbers: physical work in predictable environments, processing data, and collecting data have the highest automation potential. Ask whether interacting with people could ever join that list, and don't let them rule it out too quickly.
• "AI impact" is a different measure entirely. It was built by analysing the overlap between AI-related patents and job descriptions, and it captures anything from full automation to AI being integrated into existing work. On this measure the exposed jobs are white-collar — radiologists, lawyers — because machines have become very good at prediction and pattern recognition.

Then the inequality question. Even if automation creates more jobs than it destroys, the gains may go only to workers with technical and managerial skills — and there is evidence this has already happened in the US and Europe. The jobs growing fastest are AI- and technology-related; the ones shrinking are those with the highest automation potential.

We Have Been Here Before — Or Have We?

Mindful moment (5 min), then the history, which is the deck's strongest rhetorical move:

• 1810s — the Luddites, English textile workers who destroyed machines they feared would replace them.
• 1850 — New York City tailors strike over the sewing machine.
• 1860 — grain shovelers form a union refusing to work for employers using grain elevators.
• 1890 — England's Red Flag Act requires a person to walk ahead of every vehicle waving a red flag.
• 1930 — Keynes coins the phrase "technological unemployment."

And two cases where the fear was simply wrong:

• The ATM. Widely adopted in the late 1990s and expected to eliminate bank tellers — the number of tellers rose from 2000 to 2010. Tellers per branch fell, but ATMs made branches cheaper to open, so banks opened more of them. This is the best single example of second-order effects students will meet all year.
• The high school movement. As automation destroyed agricultural jobs around 1900, the US began requiring young people to stay in school until 16 — enormously expensive, both to build schools and in lost child labour, and one of the best investments the country ever made.

Then the honest counter-question, which the deck asks outright: but is it different this time? "People were wrong before" is not proof they're wrong now. Make students argue both halves.

Before students leave: themes, lingering questions, and one new resource for Thursday.

What Work Is For

Mindful moment (5 min), resources shared, then the week turns from forecasting to values.

• Reskilling. Displaced workers may not have the skills the new jobs need. The World Economic Forum estimates half of all employees worldwide need to upskill or reskill by 2025. Is that feasible? Who pays — companies, governments, or individuals?
• Guaranteed income. The deck reaches back to The Triple Revolution (1964), which argued that since income is tied to jobs while automation eliminates jobs, "the most important needed reform is to abolish the income-through-jobs link." Sixty years old, and it reads like it was written last week.
• Keynes's prediction. In 1930 he forecast a 15-hour work week by 2030 and an "age of leisure and of abundance." We are nearly at 2030. Why hasn't it happened — and would we want it if it did?

The closing questions: would it be good if automation removed mundane and physically punishing work? What does automation do to personal fulfilment, given what we know about the effects of job loss? Should we try to slow automation down? And what would people do if they didn't need jobs?

Score Your Participation

The four reflective questions, then the participation matrix. Students also use the participation rubric to score the week and explain the score.