PwC sorted 49,364 workers into four AI cohorts. Find your square, then move one over

PwC published its 2026 Global Workforce Hopes and Fears Survey on 29 September, built on responses from 49,364 workers across 48 countries and regions, collected in May and June. The headline is adoption: 64% of workers used AI at work in the past 12 months, ten points up on last year, and daily generative AI use rose from 14% to 22%. The more useful finding sits underneath. PwC scored every respondent on two axes: how far AI has improved the quality of their work, their creativity and skills, and their value to their employer; and how scarce their skills are in the market. It then sorted the workforce into four cohorts. Where you sit on that grid now tracks who gets training and who gets rewarded for using AI.

The grid, and what each square gets

• Front-runners, 14%: strong AI advantage, scarce skills.
• AI insurgents, 18%: skills that are less scarce, but using AI to deliver more.
• Indispensables, 11%: scarce skills, not far along the AI learning curve.
• Engine room, 56%: neither.

The gaps between the squares are wide. 51% of front-runners use generative AI daily, against 11% of the engine room, and PwC reports that most of the engine room say they have never used AI. Nearly 80% of front-runners say they can access the learning and development resources they need. Fewer than 40% of the engine room say the same.

Front-runners are nearly twice as likely as the global average to work somewhere that rewards effective AI use. Two-thirds say they are very likely to ask for a promotion or a pay rise in the next 12 months, and 29% say they are very or extremely likely to change employer in the next year. PwC's 2026 AI Jobs Barometer, cited in the same report, puts the wage premium for AI skills at 62%, up from 57% in 2025.

What the obvious reading misses

The obvious reading is that front-runners got lucky: scarce skills, a good employer, a training budget. The survey's own design says something more useful. One axis is skill scarcity, and it moves over years. The other is AI advantage, and it moves on evidence you can produce this month.

That is why the AI insurgents matter. Their skills are not scarce, and they still sit outside the engine room because they use AI to deliver more. PwC reports they have the highest weekly AI use of any cohort.

Heavy users are also the sceptical ones. 31% of daily AI users name concerns about accuracy or quality as their top barrier, six points above the global average. And 34% of daily users say AI helps them do better at the most complex and difficult parts of their job, against 29% of AI users overall. The advantage comes from knowing where the tool fails and putting it to work on the hard parts anyway.

Do not wait for the programme. Access to learning and development fell from 59% to 51% in a year, and PwC finds the engine room is less likely to be rewarded for using AI or learning new skills. PwC's advice to employers is to protect time for learning instead of stacking it on top of daily tasks. Some will. Plan as if yours will not.

One caution on the data. Daily users are more confident about job security, 68% against 57% for infrequent users. The survey shows an association. It does not show which way it runs: use may build confidence, or confident people may simply use more. Read it as a signal.

Your Next Move

1. Place yourself on the grid. Ask PwC's two questions of your own work. Has AI improved the quality of what you produce, your skills and your value to your employer, with an example you could show a manager? Would your employer find your skills hard to develop or hire? Two clear yeses puts you near the front-runner square. Two noes puts you in the engine room.

2. Move one square, on the axis you control. From the engine room, the shorter route is to the insurgent square. Pick the hardest recurring task in your role and run it with AI every week for a month, checking every output. Log quality and time, before and after. If you are an indispensable, your expertise is the scarce part: point AI at it, and record what your judgement caught that the tool missed.

3. Put the record in front of someone who sets pay. Front-runners ask for promotions because they can show the advantage. Take your month of evidence to your next one-to-one and ask what effective AI use earns in your organisation. If the answer is nothing, you have learnt which kind of employer you work for, and you now hold the evidence the 29% are taking to the market.

AI does the routine work. You do the thinking. The grid measures whether you can prove both.