Most claims about AI and jobs rest on surveys of what people expect. PwC's Global AI Jobs Barometer rests on what employers actually advertise. The 2026 edition, released on 15 June, analysed more than one billion job adverts across 27 countries and six continents, and its central finding is that the labour market is splitting into two tracks.
Key takeaways:
- Jobs requiring AI skills are growing almost eight times faster than the overall market: 69% versus 9%.
- The average wage premium for AI skills rose to 62%, up from 57% the previous year and 25% two years ago. It ranges from 118% in consumer markets to 16% in government.
- "Professionalised" roles, where AI automates routine work so human judgement matters more, are growing twice as fast as "democratised" roles, with 42% faster wage growth.
- The most AI-exposed companies grew headcount 52% since 2018 versus 36% for the least exposed. They are hiring, not just cutting.
- Skills required in the most AI-exposed jobs are changing more than twice as fast as in the least exposed.
Two tracks, defined
PwC's categories are its own, so it is worth being precise about them.
| Track | What AI does to the role | Examples PwC gives | Outcome |
|---|---|---|---|
| Professionalised | Automates the routine parts, leaving more room for expert judgement, creativity and leadership | Radiologists, recruiters | 2× job growth, 42% faster wage growth |
| Democratised | Makes the role itself easier for non-experts to perform | IT service managers, medical secretaries | Slower growth |
The logic is simple. If AI lowers the skill needed to do a job, more people can do it and its scarcity value falls. If AI removes the drudgery from a job that still needs an expert, the expert becomes more productive and more valuable.
The productivity "super-star" effect
The Barometer also tracks company performance by AI exposure. Companies in the most AI-exposed sectors recorded 34% labour-productivity growth in 2025 relative to 2018, against 24% for the least exposed. Within the exposed group, the top fifth achieved an average 163%, almost five times the group average.
Crucially, those high-productivity firms are not shrinking. Headcount at the most AI-exposed companies grew faster than at the least exposed. PwC's reading is that firms using AI well are expanding into new work rather than only cutting costs.
What employers now ask for
The report describes a shift in the skills mix towards judgement, creativity and leadership, the human capabilities that sit on top of AI rather than compete with it. In AI-exposed occupations the required skill set is turning over more than twice as fast as elsewhere, which means a qualification earned three years ago describes a different job today.
PwC says AI is "not simply automating tasks, but fundamentally redefining roles and skill requirements."
— PwC, 15 June 2026
Why learners should care
Three practical conclusions fall out of the data:
- Aim for professionalised roles. Ask whether AI makes your target job easier for anyone to do, or more powerful in the hands of someone who knows what they are doing. Prefer the second.
- AI skills are a premium, not a niche. A 62% average wage premium, rising year on year, is the strongest single salary signal in the labour market.
- Sector matters. The premium is highest where competition is fiercest (consumer markets) and lowest where it is weakest (government).
Concept to take away: complements versus substitutes. Economists distinguish technologies that replace a worker's output from those that multiply it. The same AI tool can be either, depending on the job. The Barometer is, in effect, a billion-row measurement of which jobs are on which side of that line.
Note: the Barometer uses job-ad data from Lightcast. "Professionalised" and "democratised" are PwC's own labels; define them for readers rather than assuming they are standard terms.
