37% of Workers Think AI Could Replace Them in 10 Years — The Survey Says Something More Interesting

sheen6869

4 days ago

4 min read

By Shakeel Ahmed — AI Growth Lab

In a 2026 survey of 8,234 working-age adults in four countries, 37.2% of active workers said AI or automation was certain, very likely, or somewhat likely to replace their job within ten years. That is the headline. The more interesting finding is what the same survey says workers actually value — because it explains why that fear, and the hope underneath it, are both real.

The survey, fielded by More in Common in June and July 2026 and presented in a Google DeepMind essay on October 29, 2026, covered Denmark (2,032 people), India (2,072), the United Kingdom (2,052), and the United States (2,078). Its results were explained in detail by NeoTeo.

First, what this number does — and does not — mean

It measures perceived risk, not jobs already lost. The survey asks what workers expect over the next decade. It does not count layoffs, and it does not prove AI caused any particular job loss. Confusing fear with fact is the most common mistake in this whole debate.

White-collar workers felt more at risk (40.4%) than manual and service workers (28.8%).
81% of respondents were satisfied with their jobs, even while more than a third worried about replacement.
About 20% of 7,106 active workers said they would not leave their jobs even if offered guaranteed income at 120% of their salary, with health care, pension, and paid leave preserved. That answer came from a hypothetical scenario, not a real decision — but it says something powerful about what work means beyond money.

What workers actually value

Among the 6,224 respondents classified by work values, the largest group was:

Intrinsic — 33.21%: people who value the work itself.
Transactional — 26.29%: people who value financial benefits most.
Community-oriented — 23.12%: people who value social and community aspects.
Status-driven — 17.38%: people who value social status.

The biggest group does not work mainly for money or status. They work because the work itself matters to them. That is a problem for any plan that treats workers as costs to be automated away — and an opportunity for any employer that redesigns jobs around meaningful work instead of just cheaper work.

What the wider data says

The survey's fear sits alongside a more balanced global picture. The World Economic Forum's Future of Jobs Report 2025 projects 170 million new jobs created and 92 million displaced by 2030 — a net gain of 78 million. Those are projections, not guarantees, and they describe the world, not any one person's career. But they explain why "AI will take all jobs" and "AI will create plenty of jobs" can both sound true: both things can happen at the same time, to different people.

Separate hiring evidence is also mixed. Statistics Canada's October 2026 analysis found the steepest youth hiring slowdown in jobs AI barely touches, and that coding-role hiring began falling before ChatGPT shipped — a reminder that not every hiring change is caused by AI, even when it happens at the same time.

What to do with this if you are a worker

Learn the judgment parts of your job. AI is strongest at drafting, summarizing, and routine analysis. It is weakest at decisions where context, trust, and consequences matter.
Make your work visible. Document outcomes, not just activity. The worker who can show "I solved this, for this customer, with this result" is harder to replace than the worker whose work is invisible.
Ask your employer one question: "What will my job look like in two years if AI does the routine half?" A company with no answer is telling you something important.

What to do with this if you are a boss

Do not announce AI as a headcount plan. You will get fear, not productivity — and the survey shows your best people value the work itself, not just the paycheck.
Redesign jobs before you automate tasks. Give people the interesting, human parts of the work and let AI take the repetitive parts. That is how you keep the 33% who care about the work itself.
Train in public. A training program everyone can see beats a private pilot that everyone rumors about.

The bottom line: 37.2% is not a prediction that your job will disappear. It is a warning that workers can see change coming and are not sure anyone has a plan for them. The companies that win the next decade will be the ones that answer that worry with a real plan — not with another chatbot announcement.

Sources: NeoTeo — DeepMind's Work-and-AI Survey: Findings Explained (survey by More in Common, 8,234 adults, June–July 2026; Google DeepMind essay, October 29, 2026). World Economic Forum figures via the Future of Jobs Report 2025.

Originally published on AI Growth Lab.