AI is more likely to change jobs than to remove them entirely. Routine digital tasks such as drafting, data entry and basic analysis are most exposed, while work involving judgement, relationships, physical skill and building with technology remains valuable. Students who learn to use AI well in their chosen field are best protected.
The honest short answer
Nobody can predict exactly how AI will reshape the job market, and anyone who claims certainty is guessing. What we can say with reasonable confidence is this: AI automates tasks, not whole jobs, at least at first. Most jobs are bundles of many tasks. When some of those tasks become automatic, the job changes. Sometimes fewer people are needed; sometimes the same people do more valuable work; sometimes entirely new roles appear.
Which tasks are most exposed?
Tasks are more exposed when they are digital, repetitive and based on recognisable patterns:
- Drafting routine emails, reports and descriptions.
- Data entry, copying information between systems and basic bookkeeping.
- First-level customer support for common questions.
- Simple translation and transcription.
- Basic coding tasks and boilerplate.
- Summarising documents and meetings.
Which work stays valuable?
- Judgement and responsibility: deciding what to do when the answer is unclear, and being accountable for it.
- Human relationships: teaching, caring, negotiating, leading, counselling and selling to people who want a person.
- Physical and skilled trades: many hands-on jobs are hard to automate.
- Problem framing: knowing which problem is worth solving and how to explain it.
- Building and supervising AI systems: designing, testing and improving the tools themselves.
- Deep expertise: specialists who can tell when AI output is subtly wrong.
What this means for students
The safest position is not avoiding AI. It is being the person in your field who uses it best. A junior accountant who can automate reconciliations, a marketer who can run AI-assisted campaigns, a doctor who understands AI diagnostic tools, a developer who uses coding assistants well: each is more productive and more valuable.
Practical steps:
- Keep strong fundamentals. Reading, writing, maths and subject knowledge let you judge AI output.
- Learn AI in your field. Find out how professionals in your target career are using it now.
- Build a portfolio. Show what you can make, not just what you studied.
- Develop human skills. Communication, teamwork and leadership grow in value as routine work shrinks.
- Stay adaptable. Expect to keep learning throughout your career.
A word to worried parents
Some families are questioning whether education is still worth it. It is, but the kind of education matters more than before. A child who memorises without understanding is exposed. A child who understands concepts, can learn new tools quickly and can build things is well placed. Our guide is university still worth it in the AI era? explores this further.
Where new jobs are appearing
New technology also creates work. See jobs AI will create for roles that barely existed a few years ago, from automation specialists to AI trainers and evaluators.
Frequently asked questions
Which careers are safest from AI?
No career is completely safe from change, but roles combining specialist expertise, human relationships, physical skill or responsibility for decisions are less exposed.
Should my child avoid computer science because of AI coding tools?
Not necessarily. AI changes how software is written, but people who understand systems, design and problem-solving remain in demand. The bar for shallow skills is rising.
Will AI create more jobs than it destroys?
Economists disagree, and the answer may differ by country and industry. Preparing with adaptable skills is wise either way.
