A new PwC study has put a number on something many of us in talent have sensed for a while: AI isn't affecting the job market evenly. It's splitting it in two.
According to PwC's 2026 AI Jobs Barometer, which drew on more than a billion job postings across 27 countries, roles requiring specific AI skills grew almost eight times faster than the overall job market in 2025, and the wage premium for those skills widened too from 57% to 62% year-on-year. But the more interesting finding isn't about who has AI skills. It's about how companies are choosing to use AI in the first place, and how differently that choice is playing out.
Two ways to use the same technology
PwC's report draws a clear line between two approaches. Some companies are using AI to amplify what their people are already good at: judgment, creativity, complex problem-solving. Others are using it primarily to cut costs and headcount.
The first group is winning by a wide margin. Companies with the highest AI exposure saw employment grow 52% since 2018, compared to 36% for the least-exposed firms. Productivity gains told a similar story: the top 20% of companies by AI exposure posted labour productivity gains of 163% relative to 2018 — nearly five times the average for AI-exposed companies overall.
As Joe Atkinson, PwC's global chief AI officer, put it, the companies seeing the greatest returns are the ones using AI to extend human expertise, not replace it.
The roles that are thriving look a lot like ours
The report singles out radiologists and recruiters as examples of roles where AI amplifies human skill rather than substituting for it. These roles saw job growth twice as fast, and salary growth 42% faster, than roles like IT service managers, loan officers and medical secretaries — jobs where AI mostly just makes the task easier for a non-specialist to do.
That distinction matters to us directly. Search has always depended on reading people well, understanding what a business actually needs beyond the job description, and making a judgment call on fit. AI can now help us move faster through the research and the noise.
The bigger shift: judgment is moving earlier in careers
Perhaps the most consequential finding is what's happening at entry level. Roles that require traditionally senior competencies like judgment, empathy, ethics, and leadership, have grown 35% since 2019. Entry-level roles that don't require these skills have shrunk by 10% over the same period.
PwC's global workforce leader, Pete Brown, frames this as AI quietly removing the routine work that used to function as an apprenticeship. The tasks that once let a junior employee learn the ropes by doing the easy version of the job are increasingly automated, which means the judgment and adaptability employers used to expect only from senior hires are now being asked of people much earlier in their careers.
It's a real problem for how organisations build talent pipelines, and it shows up in how CEOs are already planning. In PwC's Global CEO Survey, 49% expect AI adoption to reduce junior hiring over the next three years, against just 12% who expect the same at senior level.
What this means for how we hire
For companies across the region, the takeaway isn't to slow down AI adoption. The data doesn't support that. The most AI-exposed firms are growing headcount and productivity faster than everyone else.
The takeaway is to be deliberate about what AI is being asked to do. Is it there to make your people better at judgment-heavy work, or is it there to make an entry-level task cheap enough that you no longer need to hire for it? Those two answers lead to very different workforces five years from now — one with a stronger pipeline of people who've been trusted with real decisions early, and one with a growing gap between the junior roles that no longer exist and the senior judgment that's suddenly in short supply everywhere.
That's the conversation we're increasingly having with clients across the region: not whether to adopt AI, but what kind of organisation they're building on the other side of it.
Source: PwC 2026 AI Jobs Barometer, via The Straits Times, June 2026.
