From Farm To Factory To Office To…What?

Many people are asking themselves some form of "what does AI mean for my job?", and getting answers at very different extremes: Either AI will make us all super-powered at doing our current jobs if only we learn how, or AI is going to replace us all and leave us nothing to do. Instead of that "either-or", I think it's more helpful to see it as a "both-and." It is going to change the way we work, but it is also going to change the work we do.

Looking at a longer historical perspective shows this pattern repeating. In the early 20th century, there was a shift in work from the farm to the factory. Then in the late 20th century, there was another shift from the factory to the office. Now with AI, we're on the edge of a move from the office to...something else. But in each shift, two different changes happened in parallel.

First, the nature of existing jobs changed for the people doing them. Today's farmers do different tasks after things like tractors, fertilizers, soil chemistry, and commodities markets transformed the business of farming. A factory worker who manages automated assembly lines has a very different job from one manually tightening bolts with a wrench.

Second, the makeup of the workforce changed over time as people shifted from one kind of job to another. As each industry saw its efficiency improve with new technology and new practices, each saw its share of employment fall dramatically. Many people in those newly labor-efficient fields were pushed away from their current jobs, and pulled towards fields that still needed people to work them.

Those past transitions suggest what will happen next. First, all of the other transitions worked by using technology, automation, and management practices in conjunction to create higher productivity. These changes moved the content of the work from doing it directly by hand to leveraging new tools effectively, and we should expect that to be true for the people who remain knowledge workers in the future.

Second, each of these created so much productivity that it surpassed demand. Making farms 10x more productive didn't mean that we ate 10x more food, it meant that fewer people needed to work on farms. Workers also responded by changing jobs into fields that weren't helped by the emerging tools - factories didn't benefit from tractors, offices didn't benefit from assembly lines - and thus could absorb those workers into new jobs.

Not everyone will experience both changes the same way. Some knowledge workers will use AI's leverage to gain more agency over their work - becoming more valuable, taking on bigger problems, managing larger scopes. But many others will be pushed around by these macro forces, finding their specialized skills commoditized or their roles consolidated away, forced to navigate toward the work still needs human hands and human presence.

So what roles might not benefit from AI? One place to look: roles that require embodied, social beings. Work that's about motivating young students, supporting older people in need of care, creating meaning and connection. These are jobs where what matters is the people involved and not just the tasks completed. Interestingly, the consequence of AI might be a shift from the knowledge work office to a more hands-on, person-to-person form of work - a care economy as the new ground where automation can't compete with human labor.

The question "what does AI mean for my job?" is really two questions: How will the content of your work change as AI handles more tasks? And if your field becomes too productive for current demand, what kind of work - what fundamentally different field - could pull in you and others like you? The answers will be different for different people, but both questions are coming.