Short answer: the jobs with the most overlap with AI are writing, translation, customer service, sales and clerical work. Microsoft research found interpreters and translators most exposed, followed by roles like writers, customer service representatives and ticket agents. Exposure means AI can do many of the tasks, not that the job will disappear.
Key takeaways
- Microsoft's 2025 study ranked interpreters and translators, historians, writers and customer service reps among the most AI-applicable jobs.
- Sales, computer and math, and office support were the most exposed broad groups.
- The WEF expects the fastest declines in clerical roles such as data entry, bank tellers and cashiers.
- Exposure often means the job changes; how fast depends on the employer.
How is AI risk measured?
The Microsoft Research study of AI applicability by occupation looked at 200,000 anonymized conversations with Microsoft's Copilot and matched the tasks people used it for to the work activities of each occupation. It produced an "AI applicability" score. The authors stress this measures overlap with what AI can help with, not whether jobs will be cut.
Which jobs score highest?
- Interpreters and translators
- Historians, writers and authors, editors and proofreaders
- Customer service representatives and telephone operators
- Sales representatives of services
- Ticket agents and travel clerks
- CNC tool programmers, reporters and public relations specialists
The study found the most exposed broad groups were sales, computer and mathematical, office and administrative support, and media roles. It also found only a weak link with education: degree jobs were slightly more exposed, not less.
Which jobs are employers expecting to shrink?
The World Economic Forum Future of Jobs Report 2025 surveyed more than 1,000 employers. They expect the fastest declines in data entry clerks, bank tellers, postal clerks, cashiers and ticket clerks, administrative assistants and bookkeeping and payroll clerks, driven by digital tools and AI.
Does exposure mean you will lose your job?
Not necessarily. In many jobs AI takes over drafting, searching or routine replies, and people handle the harder cases. That can mean fewer new hires over time rather than sudden layoffs. Watch for signs in your own workplace: hiring freezes for your role, new AI tools taking over core tasks, or work being consolidated.
What should you do if your job is exposed?
- Become the person on your team who uses AI best.
- Shift toward tasks needing judgment, relationships or accountability.
- Build a skill that moves you to a growing area, such as data or project management.
- Keep your resume current and track openings in adjacent roles.
See our guides to AI-resistant jobs, starting a career in AI and beginner IT certifications.
Which tasks are most exposed?
Risk usually applies to tasks rather than whole jobs. Tasks that are often exposed include:
- Routine data entry and processing.
- Drafting standard documents and emails.
- Answering common customer questions.
- Basic translation and transcription.
- Summarising information from documents.
- Simple coding and testing tasks.
Jobs made mostly of these tasks face the biggest change. Jobs where they are a small part may simply change shape.
How can you check your own exposure?
- Write down what you do in a typical week and roughly how much time each task takes.
- Mark tasks that are routine, rule-based and done on a computer.
- Try current AI tools on a few of those tasks to see how well they perform.
- Note tasks that need judgement, relationships or physical work.
If most of your time is spent on exposed tasks, it is a good time to build new skills.
Which nearby roles could you move to?
| Exposed role | Possible next step |
|---|---|
| Data entry clerk | Data quality analyst, operations coordinator |
| Customer service agent | Customer success manager, complex case specialist |
| Junior copywriter | Content strategist, editor |
| Bookkeeping clerk | Financial analyst, accounts supervisor |
| Translator | Localisation specialist, interpreter |
What support is available?
Many employers offer training budgets and internal moves. Public employment services in many countries offer free courses and advice. In the US, CareerOneStop lists training programmes and American Job Centers. Community colleges often run short courses linked to local employers.
How should you talk about AI with your employer?
Show interest rather than fear. Ask how the company plans to use AI, offer to test tools in your team and suggest ways they could save time. People who help shape how AI is used are often better placed when roles change.
What signs suggest change is coming?
- Your company pilots AI tools in your team.
- Hiring for your role slows while hiring for AI roles grows.
- Job ads in your field start asking for AI skills.
Spotting these early gives you time to plan.
Remember that exposure is not destiny. Many workers in exposed roles will keep their jobs as tasks shift, especially those who learn new tools and take on work that needs judgement and people skills.
Start small: one new skill or tool this quarter is a realistic first step.
Frequently asked questions
Will AI replace customer service jobs?
Many routine contacts are already automated, but complex and sensitive cases still need people. Expect fewer, more skilled roles.
Are writers at risk from AI?
Writing tasks overlap heavily with AI, so writers who add reporting, expertise or editing judgment are better placed.
Are high-paid jobs safer?
Not necessarily. The research found only a small difference by wage, and degree jobs were slightly more exposed.
Will AI cause mass unemployment?
The WEF expects more jobs created than displaced worldwide by 2030, but many people will need to change roles.
Is translation still a good career?
It is among the most exposed. Specialized fields like legal, medical and live interpreting hold up better.
How can I check my own job's risk?
Look at which of your daily tasks AI tools can already do, and talk to people in your field about hiring trends.
How do I check if my job is at risk from AI?
List your weekly tasks, mark routine computer-based ones and try AI tools on them to see how much they can do.
Sources: Microsoft Research study of AI applicability by occupation; World Economic Forum Future of Jobs Report 2025.
We publish practical guidance, separate evidence from guarantees and encourage readers to verify current rules with official sources.
Keep this insight useful.
Share it with your network, save it for later, or create a signed affiliate link.
Creating or clicking a link does not create earnings. Only fraud-reviewed outcomes can allocate an existing eligible paid-ad-funded monthly pool.
About the author
Lutaf Junijo
Career Content Editor, World Careers Hub
Lutaf Junijo is the Career Content Editor at World Careers Hub and Dev Global Jobs. He writes and edits our guides on jobs, CVs, visas, salaries and careers abroad, and he runs our official social media. Every fact is checked against official and trusted sources, named in the post, and errors are corrected as soon as readers report them.
