In 34 of 37 countries, more people expect artificial intelligence to lead to fewer jobs than to more jobs over the next 20 years. That is the headline of a new Pew Research Center survey of more than 42,000 adults in 36 countries, carried out from February to May 2026, plus two U.S. surveys in February and June. These are public expectations, not forecasts of actual job losses, and they matter for anyone who teaches people to work with AI.
In the United States, about seven in ten adults expect job losses, a share that has risen 7 percentage points in two years. The numbers are similar in Australia and South Korea. In a number of countries, among them Canada, France, Singapore, Sweden and India, adults under 35 are more likely than those 50 and older to expect job losses. In many middle-income countries, people with more education and higher incomes are more likely to expect fewer jobs, partly because those with less education and lower incomes more often say they are unsure.
The survey asks about the economy, and those feelings come along into every training room. Someone who believes AI is there to replace them learns differently from someone who expects it to make their work better.
Other research shows what helps. Gallup's 2026 data on U.S. workers found that employees whose managers actively supported AI use were much more engaged: 48 percent, compared with 30 percent. In Gallup's data, manager support showed the widest gap of any factor, and access to AI alone did little for engagement.
One pattern deserves special attention. Young adults, whom many organizations count on as early adopters, are in several countries also the most likely to expect AI to cost jobs.
Talent Takeaway
Before the first AI training, ask the group what they expect the technology to mean for their jobs. Then share openly what the organization plans: which tasks change, which roles stay, what support comes with it. In our experience, people learn a new tool faster when they know where they stand.
Read the original at Pew Research Center
Sources
Silver, L., Wike, R., Fagan, M., & Lippert, J. (2026, September 17). Globally, more people expect AI to cause job loss than growth. Pew Research Center. https://www.pewresearch.org/global/2026/09/17/globally-more-people-expect-ai-to-cause-job-loss-than-growth/
Harter, J. (2026, July 21). Employee engagement remains flat as AI adoption accelerates. Gallup. https://www.gallup.com/workplace/712433/employee-engagement-remains-flat-adoption-accelerates.aspx