PwC: AI Will Raise Pay in Some Jobs, Cut It in Others

Only three in ten workers in Europe can be certain that the growth of AI-assisted automation will not significantly affect their job. That's according to the latest edition of the PwC AI Jobs Barometer report, which, instead of a simple split between "at-risk" and "safe" professions, reveals a deeper mechanism: AI is splitting the labor market into two paths, one leading to higher pay, the other to growing competition for lower wages.

Two paths for the labor market

The authors of the PwC report describe two opposing trends that AI is triggering simultaneously in different parts of the economy. The first is democratization: artificial intelligence takes over routine, repetitive elements of work, so tasks that once required years of experience can now be handled by people earlier in their careers. That sounds like a convenience, but it has a flip side: if more people can do a given job, its market price falls.

The second trend is professionalization. In some occupations, AI doesn't replace people, it raises the bar - clients and employers expect more than task completion, they want experience, judgment and soft skills that language models can't replicate. It's in this group of occupations that the report records the fastest wage growth and the most dynamic expansion of required competencies.

Who loses, who gains

Among the occupations the report flags as susceptible to democratization are interior designers, administrative staff, accountants, call center consultants, systems administrators, IT project managers and programmers. In these roles, AI is taking over a growing share of routine tasks, which, according to the report's authors, lowers the entry barrier but also increases competitive pressure on wages.

On the other side are occupations tied to managing people and creating things: HR specialists, recruiters, career advisors, creatives (musicians, songwriters), environmental engineers, R&D managers and corporate secretaries. According to PwC, these roles are gaining importance precisely because they require empathy, communication and creativity, qualities that are hard to automate.

Not job losses, but a reshuffle

The report's key conclusion diverges from the most alarmist forecasts of mass layoffs. PwC stresses that for most workers, AI doesn't mean losing employment, but a change in the nature of the work performed and, more importantly from a paycheck perspective, a shift in the relationship between the supply of skills and their market price.

Artificial intelligence itself won't take your job, someone who knows how to use it better than you will.

What it means for workers

The report's authors recommend two strategies for people working in occupations most susceptible to democratization. The first is retraining toward professionalized roles, where both wages and demand for experience are rising. The second is considering a career in occupations with low exposure to AI competition, for example in skilled trades or occupations requiring physical work and specialized licenses.

For programmers, accountants or systems administrators, occupations explicitly named in the report as susceptible to democratization, the conclusion is uncomfortable: simply having a technical skill is no longer enough to maintain one's previous negotiating position in the labor market if the same skill, aided by AI tools, can be offered by far more candidates.

Implications for the Polish market

PwC's report covers the European market, but the mechanism it describes has a direct bearing on Poland, where the IT and business services sector (BPO/SSC) employs hundreds of thousands of people in administrative, accounting and programming roles, exactly the categories PwC classifies as most exposed to democratization pressure. At the same time, Poland is also where the number of companies seeking soft and managerial skills is growing, which aligns with the professionalized segment of the market identified in the report.

PwC's conclusions don't suggest that Polish programmers or accountants will lose their jobs overnight. Rather, they suggest that over the next few years, these workers' pay and negotiating position will increasingly depend on whether they can shift toward tasks requiring judgment, relationship management and integrating the work of AI tools, rather than simply performing tasks these tools can already do on their own.