Updated 7 October 2026. A headline about AI and jobs can mix three different things: what software can do, what an employer chooses to automate and how employment actually changes. They are not interchangeable. This guide explains how to assess those claims and plan your learning.
The earlier version used unverified layoff totals, company statements and anonymous “expert” predictions. Those have been removed. This page is not a live layoff tracker or a forecast of how many jobs AI will create.
Exposure is not a job-loss count
The ILO–NASK study announced in May 2025 estimated that one in four jobs worldwide had some potential exposure to generative AI. It identified transformation as more likely than complete replacement. This was a task-based exposure assessment, not a count of workers already dismissed because of AI.
A role can include drafting, customer communication, checking, judgement and physical tasks. Automating one part does not by itself prove that the whole role disappears. Equally, exposure research does not guarantee that every worker benefits.
Check a layoff claim before repeating it
- Find the employer’s dated announcement or filing.
- Check whether the figure is announced positions, completed departures or an estimate.
- Separate global totals from a country, office or business unit.
- Read the stated reasons; do not assign every reduction to AI.
- Look for the counting period and whether separate reports may count the same workers twice.
If those details are absent, treat a precise total as unverified. This guide does not retain the earlier 92,000 figure or attribute a specific number of new jobs to a technology executive.
Build a skills plan from actual vacancies
- Collect a small set of current listings for the role and location you want.
- Separate essential requirements from preferred skills. Record experience and qualification requirements as written.
- Find repeated tasks rather than chasing an impressive job title.
- Choose one gap you can demonstrate through a project.
- Build a sample, document its limitations and practise explaining your decisions.
For example, someone exploring an analyst role could create a small report from public data, show how it was cleaned and explain uncertainty. Someone exploring customer support could practise clear responses and escalation decisions. These are portfolio ideas, not documented hiring results or guaranteed routes into a role.
Use AI without concealing your contribution
If a tool helps with a project, record where you used it and what you checked. Do not claim independent authorship of work you cannot explain. Do not upload an employer’s or client’s private data without permission.
Learn the underlying task as well as the interface. A prompt alone does not demonstrate subject knowledge, quality control or the ability to fix an incorrect result.
What the IMF’s 2026 research means for job seekers worldwide
Research update — 19 October 2026. The International Monetary Fund distinguishes AI exposure from actual job losses. Its February 2026 overview estimates that AI may affect roughly 40% of jobs worldwide and 60% in advanced economies. “Affected” includes work that may be improved, reorganised or displaced; it is not a forecast that those percentages of workers will lose their jobs.
The IMF’s January 2026 skills analysis reports that about one in ten job postings in advanced economies and one in twenty in emerging-market economies request at least one new skill. Those figures cover emerging skills more broadly, not exclusively AI skills. They also describe the analysed postings, not every vacancy in every country.
This IMF analysis and the ILO–NASK generative-AI estimate discussed above use different definitions and methods. Do not add their percentages together or interpret the difference as a sudden change in layoffs. For a practical next step, compare ten recent vacancies for the same role in your own country: note which tasks, digital skills and experience requirements repeat. Select one demonstrable project based on that evidence, then review vacancies again in a month. Training may help you prepare, but it does not establish that an employer will hire you.
For Indian job seekers
Check the location, work arrangement, employment type and eligibility in each actual listing. Do not assume a course certificate replaces a mandatory degree or that a remote listing accepts applicants from every country.
Before paying for training, inspect the syllabus, sample lessons, total fee and written refund terms. Ask how any placement statistic was calculated. A skills course cannot guarantee a job merely because it includes AI.
Correction: unsupported statistics and attributed predictions have been replaced with the scope and limits of an identified study and a practical verification process. No salary or job-placement outcome is promised.
To turn a broad career concern into a role comparison, see technology career options and pay-offer checks; compare actual requirements rather than assuming exposure predicts your own outcome.
