The Double-Edged Sword of Technological Progress
Every technological revolution tells the same story: machines replace some jobs while creating entirely new categories of work. Tractors replaced farm hands, but enabled aeronautical engineers to exist. As we stand at the threshold of the AI revolution, a crucial question emerges: who benefits when technology creates new opportunities?
A groundbreaking study led by MIT labor economist David Autor provides fascinating insights into this question, revealing patterns that could help us understand AI's potential impact on the job market.
Young, Educated, and Urban: The Winners of Technological Change
Autor's research, published in the Annual Review of Economics, analyzed U.S. employment data spanning from 1940 to 2023. The findings are striking: new forms of work have consistently benefited college graduates under 30 more than any other demographic.
"We had never before seen exactly who is doing new work," Autor explains. "It's done more by young and educated people, in urban settings."
The numbers tell a compelling story:
- In 1950, about 7% of employees worked in job categories that had emerged since 1930
- From 2011-2023, roughly 18% of workers were in roles introduced since 1970
- College graduates were 2.9 percentage points more likely than high school graduates to land new work
- Workers under 30 consistently outpaced other age groups in accessing emerging job categories
The Economics of Expertise: Why New Work Pays More (At First)
New work comes with a crucial advantage: a wage premium. These emerging roles typically offer better salaries than established positions, but there's a catch. As Autor notes, "If everyone is an expert, then no one is an expert."
This scarcity principle drives the economic value of new work. Consider how driving a car was once specialized expertise, or how mastering WordPerfect in the 1990s could boost your career prospects. Over time, these skills became commonplace, and their premium value evaporated.
"The scarcity value erodes," Autor observes. "It becomes common knowledge. It itself gets automated. New work gets old."
Demand-Driven Innovation: Lessons from World War II
One of the study's most intriguing findings challenges how we think about innovation. Rather than being purely supply-driven (entrepreneurs creating new products), much technological progress is actually demand-driven.
The research shows that government-backed expansion during World War II created massive amounts of new work and expertise. Counties with new factories had more new work, and 85-90% of new jobs from 1940-1950 were technology-driven.
"This says that wherever we make new investments, we end up getting new specializations," Autor explains. "If you create a large-scale activity, there's always going to be an opportunity for new specialized knowledge that's relevant for it."
What This Means for the AI Revolution
So where does this leave us with artificial intelligence? Autor is cautiously optimistic but emphasizes that outcomes aren't predetermined—they depend on our choices.
"People are really worried that AI-based automation is going to erode specific tasks more rapidly," he notes. "But we don't know what [new work] will be, what it will look like, and who will be able to do it."
The healthcare sector illustrates this perfectly. AI could be implemented in two very different ways:
- Job replacement: Simply automating existing roles away
- Job augmentation: Enabling workers with different skill levels to tackle new tasks and responsibilities
"I would say the latter is more socially beneficial," Autor argues. "But it's not clear that is where the market will go."
The Path Forward: Policy and Purpose Matter
Here's the key insight for prompt engineers, AI enthusiasts, and technologists: innovation is purposive, not accidental. The direction of AI development—whether it creates good jobs or eliminates them—will largely depend on conscious decisions by companies, governments, and society.
With over half of U.S. healthcare dollars being public funds, Autor points out that "we have a lot of leverage there, we can push things in that direction."
For those working with AI and prompts today, this research suggests a few important considerations:
- Stay adaptable and continue learning—expertise advantages are temporary
- Focus on developing skills that complement rather than compete with AI
- Consider how your work could augment human capabilities rather than replace them
- Remember that demand-driven innovation often creates more sustainable job growth
As we navigate the AI revolution, history suggests that young, educated professionals will likely capture many of the initial benefits. But the ultimate impact—whether AI creates broadly beneficial new work or concentrates gains among a few—will depend on the choices we make today.
This research was conducted by David Autor (MIT), Caroline Chin (MIT), Anna M. Salomons (Tilburg University), and Bryan Seegmiller (Northwestern University), with support from multiple foundations including the Hewlett Foundation and Google Technology and Society Visiting Fellows Program.