Key takeaways

• PwC’s 2026 Global AI Jobs Barometer — built on more than a billion job adverts across 27 countries — puts the average wage premium for AI-skilled workers at 62%, up from 57% a year ago.
• The labour market is splitting into two tracks. “Professionalised” roles, where AI amplifies expert judgement, are growing at twice the rate of roles AI merely simplifies — and their salaries are rising 42% faster.
• In the UK, specialist AI postings grew 61% in a single year (112,000 to 180,000) while vacancies across the whole economy fell 6.6%. The UK’s AI wage premium tripled, from 11% to 34.2%.
• Health accounts for less than 1% of AI job growth — the starkest gap in the entire sector table, and the clearest arbitrage for organisations willing to move before their competitors do.
• AI-exposed entry-level roles are now seven times more likely to demand senior-level judgement. The apprenticeship route into AI capability is closing — which makes the experienced leadership hire the load-bearing decision of the next eighteen months.

The report every board pack will quote this quarter

Every July, PwC’s Global AI Jobs Barometer gives the AI labour market its annual physical. The 2026 edition, published this month, is the largest yet: over a billion job advertisements analysed across 27 countries and six continents. It is about as close to a census of AI hiring as exists, which is why its numbers will be circulating in board packs for the rest of the year.

The headline is easy to quote and easy to misread. AI-skilled workers now command an average wage premium of 62%, up from 57% last year. AI-specific roles grew 69% against a broader job market that managed 9%. In the UK, specialist AI postings rose 61% in a year in which overall vacancies fell 6.6%. As PwC’s UK Chief Technology and Innovation Officer Claire Reid put it: “The experimentation phase is over and businesses want to scale and embed the technology properly.”

But the number that should actually change behaviour sits further down the sector table, and almost nobody will lead with it. We will, because it concerns the sectors we serve: health accounts for less than 1% of AI job growth.

A market splitting in two

The Barometer’s most useful contribution this year is a distinction between two kinds of AI-touched work.

In professionalised roles, AI amplifies expert judgement — the clinician who reads model output against twenty years of pattern recognition, the data science leader who knows which validation failure matters and which is noise. In democratised roles, AI simply makes tasks easier for non-experts.

The two tracks are diverging fast. Professionalised roles are growing at twice the rate of democratised ones, with salaries rising 42% faster. Companies most exposed to AI grew headcount 52% since 2018 against 36% for everyone else — and the top fifth of those AI-exposed companies posted labour productivity gains of 163%, nearly five times the broader cohort.

“Across the global economy, we’re beginning to see a new divide emerge between different models for talent and value creation,” is how Joe Atkinson, PwC’s Global Chief AI Officer, frames it. The divide is not between companies that use AI and companies that don’t. It is between organisations that hired people capable of compounding AI into expertise — and organisations that bought licences.

The counter-intuitive part

Here is what the premium data does not say: it does not say the money is in technology companies. The largest wage premiums appear where AI skills are scarcest relative to demand — consumer markets pay a 118% premium globally precisely because so few AI-fluent operators work there. Scarcity, not sector glamour, sets the price.

The UK tells the same story from a different angle. A 34.2% national premium — triple last year’s 11% — emerged while the wider vacancy market contracted. Employers are not paying more for AI skills because hiring is booming. They are paying more because these are the hires they refuse to do without, in a year when everything else got cut.

For healthcare and life sciences organisations, that logic cuts both ways. The premium you pay reflects the scarcity you face — but it also predicts the premium your competitors will face twelve months from now, when the same conclusion reaches their board. We covered the mechanics of this escalation in The AI Compensation Arms Race; the 2026 Barometer is that argument with a billion adverts behind it.

Healthcare’s one percent problem

Technology, media and telecommunications take 11% of AI job growth. Professional services take 6%. Health takes less than one percent.

Read one way, that number is an indictment: the sector with the highest-stakes use cases — diagnostics, clinical NLP, drug discovery, patient-risk prediction — is barely staffing the capability. The workloads are arriving regardless: regulatory deadlines like the EU AI Act’s August 2026 obligations (which we mapped here) do not wait for headcount plans, and the organisations deploying clinical AI without senior AI leadership are accumulating risk they cannot see.

Read the other way, it is the most attractive arbitrage in the table. Elsewhere, AI leaders are bid for by ten rivals at once; in healthcare and life sciences the internal competition has not yet woken up. The organisations building their healthcare AI leadership team now are hiring into the last quiet market they will see this decade. On current trajectory, health’s sub-1% share converges toward the professional-services number within a couple of reporting cycles — and the premium converges with it.

The apprenticeship is disappearing

One more finding deserves a hiring leader’s attention. Analysing 2.4 million US entry-level roles, PwC found that AI-exposed junior positions are now seven times more likely to require senior-level capabilities — judgement, leadership, face-to-face persuasion. “Seniorised” entry-level roles have grown 35% since 2019; the rest have shrunk 10%. As PwC’s Global Workforce Leader Pete Brown observes: “AI is removing some of the routine work that once acted as an apprenticeship, while increasing demand for judgement, leadership and adaptability much earlier in careers.”

The comfortable plan — hire bright juniors, grow your AI capability organically, avoid the senior premium — is quietly losing its supply chain. The routine work those juniors would have learned on is precisely the work AI now does. Capability increasingly has to be hired in at the top and taught downward: an experienced leader who has taken models from pilot to production, building judgement in the team faster than the market seniorises around it.

That inverts the usual sequencing question. The first AI hire is no longer the person who builds the first model. It is the person who can stand between a 62%-premium market and your board, and make every subsequent hire cheaper, faster and less likely to fail.

What hiring leaders should do now

1. Re-price your open roles against the new data. If your compensation bands predate this report, they predate a 62% global premium and a UK premium that tripled in a year. Benchmarks from January are already historical documents.

2. Classify your roles honestly. Which of your open positions are professionalised — amplifying judgement you genuinely hold — and which are democratised work that AI will absorb? Fund the first category properly and stop over-speccing the second.

3. Hire the multiplier before the multiplied. One production-tested AI leader changes the economics of every downstream hire. Our executive search work in healthcare and life sciences AI exists for exactly this decision.

4. Rebuild the apprenticeship you just lost. If entry-level roles no longer teach, your seniors must. Weight leadership hires toward people with a record of developing teams, not just shipping models — then let structured talent acquisition fill the layers beneath them.

5. Move while your sector is still asleep. Health’s one percent is a closing window, not a permanent discount. The premium you defer paying this year is the larger premium you will pay against ten competing offers next year.

The 2026 Barometer will be remembered for confirming the divide everyone suspected. The organisations it will reward are the ones that read the sector table to the bottom — and noticed that in healthcare and life sciences, the divide has not finished forming yet.

Fergal Nolan is the founder of Banba, a specialist executive search firm for Healthcare AI and Life Sciences AI, with offices in New York, London and Berlin. He has spent more than 25 years in talent acquisition, including as a founding team member at SThree / Real Staffing and as founder of Upstream, where he has built Data Science & AI teams and AI leaders for start-ups, scale-ups and global enterprises.