Will AI take jobs?

Why people are worried

AI tools can now write text, summarise documents, answer customer questions, produce images and help write computer code. These are tasks that people are paid to do, so it is natural to wonder: will my job still exist?

The honest answer is that nobody knows exactly. Forecasts about the future of work have often been wrong in both directions. But we can look at what research says so far, and at what has happened with big technology changes in the past.

Jobs are bundles of tasks

A useful way to think about this: a job is not one thing, it is a bundle of tasks. A receptionist answers calls, books appointments, greets visitors, solves problems and calms upset customers. AI might take over the booking, but greeting people and handling a tricky situation still need a human.

So the more helpful question is not "will AI take my job?" but "which parts of my job might AI do, and what will I spend my time on instead?"

What the research says

In May 2025, the International Labour Organization (ILO), a United Nations agency, published a study with Poland's national research institute NASK. It estimated that about one in four jobs worldwide are in occupations with some exposure to generative AI, with a higher share in richer countries. Clerical and office support work was the most exposed group.

Crucially, the researchers concluded that for most of these jobs, transformation is more likely than replacement. Many tasks can be done faster with AI, but still need a person to check, decide and take responsibility. They also stressed that "exposure" is not the same as job losses, and that results will depend a lot on how governments and employers manage the change.

The study also found that women are more often in the most exposed office roles, which makes training and support especially important.

Lessons from earlier technology

We have been here before. Cash machines did not end bank jobs overnight; staff moved towards advice and customer service. Spreadsheets changed the work of bookkeepers and accountants, and created new roles in data and analysis. The internet removed some jobs, such as many in printed directories, and created many others that did not exist before.

Change can still be painful for the people in the roles that shrink. Some workers will need to retrain, and that is easier with support from employers and governments. Being honest about that is part of a balanced view.

Jobs that may grow

Alongside change, AI is creating demand for people who can:

• Use AI tools well in everyday office work and check their output.
• Look after the computer systems, networks and data centres that AI runs on.
• Keep systems secure, because criminals use AI too. Cyber security remains in demand.
• Explain technology to others: trainers, support staff and teachers.
• Do work that depends on people: health and social care, skilled trades, hospitality and teaching.

What you can do now

1. Try the tools. Spend a little time with an AI assistant on safe, everyday tasks. Our guide to AI for everyday tasks has ideas.
2. List your tasks. Write down what you do in a typical week. Mark which tasks are routine and which need judgement, care or skill. Grow the second group.
3. Learn alongside AI. Being the person who knows how to use AI well, and safely, is valuable in almost any workplace. Read using AI tools safely.
4. Keep learning. Short courses, free online learning and workplace training all count. Technical skills such as networking and cloud are a good path; see what is a network if you are curious.
5. Talk to your employer about how they plan to use AI and what training they offer.

Tip: Be cautious of headlines with dramatic numbers. Check who made the claim, what they measured and whether they mean "jobs lost" or just "tasks affected".