09 Jul 2026

When AI raises the bar, who pays the price?

With AI still at its early stages, most employers have not had time to properly weigh up any consequences it has for their staff.

Artificial intelligence now shapes daily work by speeding up decisions, closing skill gaps and changing what employers expect from their teams. The opportunities are real, but so are the risks. Most companies, however, are only paying attention to the benefits.

While AI is creating real value, it is also putting pressure on employees who feel they need to keep up with expectations. That’s why, if AI strategies are not managed properly, organisations could lose the very people they rely on to get results.

The market for AI talent is already getting tighter

AI skills have become much more valuable. Employees who use AI at work now earn 56% more than those in similar jobs who do not, and this pay gap grew by 25% in just one year. While total job postings fell by 11.3%, jobs requiring AI skills rose by 7.5%, according to PwC's Global AI Jobs Barometer 2025.

Overall, the job market seems to be shrinking, but competition for AI talent is getting tougher.

Keeping employees is the biggest challenge for HR leaders. Wellhub’s latest Return on Wellbeing study found that 62% of HR professionals worry about losing staff with AI or related skills, and this rises to 64% in small and medium-sized businesses. In jobs most affected by AI, skill needs are changing 66% faster than in other areas. For people in these roles, keeping up is not optional. It is necessary to stay relevant.

Reward systems based on yearly benchmarks were not made for such rapid change. If pay and career progression do not keep up with the market, organisations risk losing the very people they need most.

'Shadow AI' is dividing the workforce

As companies expect more AI skills, most employees are still learning how to use it. McKinsey’s The State of AI in 2025: Agents, innovation and transformation report found 88% use AI in some part of their business, but only 7% have rolled it out across the whole company. This gap between trying out AI and fully using it as a formal process is called “shadow AI”.

For example, 78% of employees who use AI at work bring their own tools, ie they choose their personal preference of AI to use rather than a company approved AI. While 53% are unsure whether they should tell anyone when they use AI for important tasks. The technology is being used, but the rules and oversight are not keeping pace.

This has created a workforce divided in ways that are not about seniority or job title. Some employees got early access to AI, automated parts of their work, and now work at a different pace. Others have not made this change because they do not know how to start, when to use it, or do not trust in its real impact.

This divide shows who started using AI first, not who is most skilled.

There is also a psychological cost that does not show up in productivity data. More than half of workers worry that using AI for important tasks will make them seem replaceable. This fear makes people less open and less willing to share what they learn, which could help companies close the gap.

Productivity and wellbeing in the same conversation

Companies that get the best early results from AI do not treat technology adoption and people strategy as separate efforts. This approach matters because the risks of mistakes are real.

When top performers start using AI tools first and their output increases, expectations often rise for the whole team, even before everyone else has the same access or training. Researchers call this performance compression: higher standards are set unevenly, and the extra pressure falls on those already doing the most. In this scenario, 68% of workers say they are struggling to keep up with the pace of work, and 46% feel exhausted.

Investing in wellbeing is not just a solution after problems appear. It is the foundation that supports sustained strong performance. our Return on Wellbeing (2026) report found that, 91% of organisations with wellbeing programmes report better productivity, and 82% say these programmes are key to keeping their best people.

For reward and benefits professionals, the message to leaders is clear. Using AI without supporting the people who use it may bring quick gains but it risks losing valuable employees over time. The organisations that stay ahead will be those that treat AI adoption and employee wellbeing as part of a single strategy.