Will AI create jobs, or destroy them? Everyone wants a clean answer. The honest truth is simple. Nobody knows the exact net number. But the major labor forecasts do tell us one thing. They show the shape of the change. And the shape matters more than the net.

What the major forecasts say

The World Economic Forum's Future of Jobs Report 2023 looked ahead to 2027. It projected that AI and related tech would create about 69 million new jobs. It would also wipe out 83 million. That is a net drop of about 14 million jobs. That equals roughly 2% of jobs today. Not the apocalypse. Not nothing.

The World Economic Forum's Future of Jobs Report 2025 came out in January 2025. It looked ahead to 2030. It projected 170 million new jobs created and 92 million lost. That is a net gain of 78 million jobs worldwide. The framing shifted toward net growth. The drivers were green energy, tech roles, and the care economy, plus AI.

So the two top forecasts came out within two years. Yet they gave very different net numbers. That should tell you something. The net number is truly uncertain. It leans heavily on the assumptions you pick. One thing stays the same across both. The churn is huge. Tens of millions of jobs shift on both sides.

The historical pattern

Every big tech wave has followed the same pattern. It destroys some jobs and creates new ones. Usually it creates more in total. But the switch is painful for people in the lost roles. The car killed the horse-carriage trade. It built the auto industry instead. The internet killed travel agencies. It created whole new sectors. The lost jobs were real. The new jobs went to different people.

AI is likely to follow this pattern. The total number of jobs may well grow. But that is cold comfort to one person. Say you are a data-entry clerk. Your role gets automated. The new jobs are AI ops roles. And you are not ready for them.

Why "net jobs" is the wrong thing to focus on

Here is the deeper lesson. The net number is an abstraction. It does not help any one person. What helps a person is understanding the shift. Which jobs are growing? Which are shrinking? And how do you move from one to the other?

Take production teams at agencies that add AI to their workflows. The total headcount often does not change much. But the mix of roles changes completely. Teams need more people who can direct AI tools and check quality. They need fewer people doing purely manual work. To those workers, the net staying flat means little. What matters is one thing. Can they move into the new kind of role?

The jobs being created

AI is creating jobs that barely existed a few years ago:

The jobs being eliminated

It is also wiping out jobs that were stable for decades. Think data entry and basic content production. Think routine customer service and simple document processing. Think entry-level coding tasks. Pew Research documented the pattern in 2023. The risk lands hardest on white-collar pattern-matching work.

The honest take

Will AI create more jobs than it destroys? Probably yes, in total, over a long enough span. That is what most forecasts and history suggest. But that total answer is almost useless to you as one person. The new jobs will not flow to the people who lost old ones. So the real question is not "will there be enough jobs?" It is "will I be ready for the jobs that exist?" That part, you control.

Sources

Related reading: What Jobs in Your Field Will AI Automate First? · What's the Easiest AI Job to Get Into? An Honest Ranking