The gap between companies that trained their people and companies that just bought licenses is widening every quarter, not closing. PwC's 2026 AI Jobs Barometer (1 billion+ job ads analyzed across 27 countries) put numbers on something I keep seeing across my client base.
This isn't about which AI tool a company picked. PwC frames these as "professionalised vs. democratised" roles, and the divide comes down to whether AI is amplifying someone who already knows the work, or just lowering the bar for anyone to attempt it. Same technology, opposite trajectory.
The numbers show up everywhere:
Headcount growth: 52% at the most AI-exposed companies, vs. 36% at the least
Wage growth: 24% vs. 17%, same split
Wage premium for AI skills: 62%, up from 57% last year.
Professionalised roles grow twice as fast and pay more, because AI amplifies expertise that's already there. Democratised roles grow slower with smaller pay gains, because AI just lowers the bar for anyone to attempt the task.
The mistake I see most: "we bought the license, we're covered." The license was never the variable. Training and structure around it is.
The entry-level finding is the one I keep coming back to: entry-level roles most exposed to AI are now seven times more likely to require senior-level skills like judgement and leadership and postings for these roles have grown 35% since 2019, while other entry-level roles shrank 10%. That's not a hiring problem companies can solve by being pickier.
It's a TRAINING PROBLEM and it starts before someone's first day!
If your team is stuck on this exact question, that's what I help solve.
Source: PwC 2026 Global AI Jobs Barometer
Here is the link: https://www.pwc.com/gx/en/news-room/press-releases/2026/pwc-2026-ai-jobs-barometer.html