The job with no name

Try explaining a social-media manager to someone in 1990. The role sounds obvious now; then, the internet had not yet made it ordinary. New tools do create work people could not have pictured in advance. That history is why the hopeful AI prediction feels plausible: perhaps the jobs of tomorrow have not been invented yet.

I hope they arrive. But saying “new jobs will appear” can become a way to wave away the person whose job changes this year. A future role does not pay this month’s rent. And the new work may ask for skills, location, credentials, or experience that the displaced worker does not have. “The economy will adapt” is a description of a process, not a promise that every person lands safely.

Exposure is not a pink slip

The International Labour Organization estimates that about one in four workers worldwide is in an occupation with some exposure to generative AI. Exposure means that parts of a job may be affected. It does not mean one in four workers will be replaced. The ILO’s conclusion is that many jobs are more likely to be changed than eliminated.

That distinction is worth holding onto. A tool can take over a task and still leave the job in place. A lawyer may use software to summarize documents but remain responsible for judgment. A customer-service worker may get suggested replies while still handling the conversation. The work changes, sometimes in ways that help, sometimes in ways that increase pace or monitoring.

There are also jobs where tasks can be automated faster than new duties arrive. Workers may find their hours cut, hiring slowed, or entry-level work reduced, even before anyone announces dramatic layoffs. People can lose an opportunity without a robot standing beside them on the day it disappears.

Why the first rung matters

A Stanford study tracking employment data in the United States reported that, by June 2026, employment among workers ages 22 to 25 in occupations highly exposed to AI was about 19 percent below the path seen among less-exposed peers. That is a relative shortfall for a specific age group and set of occupations. It does not mean AI fired 19 percent of young workers. The researchers describe an observed pattern, not proof that AI caused every difference. Weaker hiring appears more important than a surge in layoffs.

Even with that caution, the pattern raises a real question. Many careers begin with tasks that are repetitive: draft the first version, check the numbers, answer the basic question, organize the file. Those tasks may be exactly what new systems can do quickly. But they are also how a beginner learns the work, earns trust, and gets a chance at harder assignments.

If companies remove the entry-level rung and later say they cannot find experienced people, that is not a mysterious labor shortage. It may be a ladder with the bottom missing. The concern is not that every starter job must remain unchanged forever. It is that employers need to build new ways for people to learn, practice, and prove they can do more.

A new door is not enough

The history of technology includes both new work and real losses. Online shopping created jobs in software, logistics, and fulfillment; research has also found pressure on some nearby traditional retailers. Both can be true at once. People who benefited from one side of the change do not cancel out the people hurt by another.

AI could create new services, teams, and specialties. We cannot yet say how many, how quickly, or who will be able to reach them. We should be curious about what becomes possible and skeptical of anyone selling certainty—whether the certainty is “every job is safe” or “no one will work again.”

Learning to use these tools is sensible. So is paying attention to the shape of the work around us: are people gaining time to do better work, or are they simply expected to produce more with fewer colleagues? Are companies training junior workers, or quietly outsourcing the training to the one person who survives?

The future may contain jobs we cannot name. I want those doors to open. I also want us to notice whether the people at the beginning of a career can still find the stairs.