AI may create more jobs than it eliminates. The outcome depends on how leaders use the capacity it creates: cost savings, heavier workloads, or greater value.
by Janet Ply, PhD · The Practical Leadership Newsletter · September 15, 2026
Imagine that AI saves every member of your 20-person team five hours a week.
You have just gained 100 hours of capacity. That is roughly the equivalent of adding two and a half employees without hiring anyone.
What happens next?
Do you eliminate two positions?
Expect everyone to produce 25% more?
Or use that capacity to solve problems, improve customer service, develop people, and pursue opportunities no one previously had time to address?
We spend a lot of time debating whether AI will eliminate jobs. It certainly will eliminate some. Other jobs will shrink or change significantly.
But that is only half the question.
AI may also create jobs, expand existing roles, and enable people to perform more valuable work. We cannot know the final employment numbers yet, but history and emerging research give us good reason to question the assumption that AI will inevitably leave us with fewer jobs.
We know that AI creates capacity for those who use it properly. The immediate question is what leaders will do with this excess capacity.
Every major technological shift has eliminated certain kinds of work.
The internet disrupted entire industries. It also helped create web development, e-commerce, cybersecurity, digital marketing, cloud computing, online education, app development, and the creator economy.
Few people at the beginning of the internet era could have predicted all those jobs. We may be in a similar position with AI.
According to Goldman Sachs Research, approximately 60% of American workers are now employed in occupations that didn’t exist in 1940. Its researchers estimate that technology-driven creation of new positions accounts for more than 85% of employment growth since then.
That does not prove AI will produce the same result. AI can perform cognitive tasks that previous technologies could not, and it is developing rapidly.
But history should make us skeptical of confident predictions that technology will permanently eliminate our need for human work.
A job is not one activity. It is a collection of responsibilities requiring different combinations of knowledge, judgment, communication, relationships, and physical action.
AI may summarize a report without understanding the political dynamics behind it. It may analyze data without recognizing which assumptions deserve to be challenged. It may draft feedback for an employee without taking responsibility for the conversation or its consequences.
That is why saying “AI can perform 30% of someone’s tasks” is not the same as saying “AI can replace 30% of the employees.”
The International Labour Organization’s 2025 research found that one in four jobs worldwide has some exposure to generative AI. But exposure does not mean elimination. Because most occupations still contain tasks requiring human involvement, the researchers concluded that job transformation is the more likely outcome.
One of the most important workplace studies involving AI followed 5,172 customer-support agents.
Researchers Erik Brynjolfsson, Danielle Li, and Lindsey Raymond found that AI assistance increased productivity by approximately 15%.
The average improvement was significant. But the most interesting finding was who improved the most.
The greatest gains occurred among less-experienced and lower-performing employees. The AI assistant helped them apply practices associated with more successful colleagues and move up the learning curve faster.
The technology did not simply make the best employees faster. It helped other employees become more capable.
MIT economist David Autor believes this effect could extend well beyond customer service. In “AI Could Actually Help Rebuild the Middle Class,” he argues that AI could make valuable expertise available to more workers.
It could help nurses perform more sophisticated care-management work, enable technicians to diagnose complex equipment problems, or allow less-experienced professionals to make better decisions with appropriate guidance and oversight.
AI can reduce the time required to complete existing work. It can also make work possible that previously required more experience, time, money, or specialized assistance.
That creates the potential for new services, new businesses, new markets, and new jobs.
AI does not automatically free people for higher-value work. It merely frees time.
Leaders decide what happens to it.
If organizations use AI-created capacity primarily to reduce headcount, the result may be fewer jobs.
If they maintain every existing priority and simply increase output expectations, employees may end up doing more work at a faster pace. Three people will be expected to do what five people did before.
But if leaders use the capacity to improve service, solve neglected problems, develop new offerings, strengthen decisions, and pursue growth, AI may create more valuable work and potentially more jobs.
Most organizations will probably do some of each. But the balance they choose matters.
My concern is that leaders will make a series of reasonable-looking, short-term cost decisions without considering the workforce they are creating or the capabilities they are eliminating.
Before asking, “How many positions can AI eliminate?” leaders should ask:
These questions apply to leaders at every level.
You may not control your organization’s workforce strategy, but you influence how AI is used within your team. You can use it to remove mindless work, spread expertise for cross-training, improve decisions, and address important work that has been neglected.
AI may create the capacity, but leaders determine whether that capacity becomes cost savings, heavier workloads, or greater value.
The future of work is not something leaders are simply waiting to discover.
Through the choices they make about automation, augmentation, employee development, and growth, they are helping create it.
So, return to that 20-person team.
AI has just given you 100 additional hours of capacity every week.
What would your organization do with it?
Janet Ply, PhD
Author, Practical Leadership ~ Creator, Leadarity AI Thinking Partner and the Practical Leadership System