13th May 2026
Every boardroom conversation about AI concerns the same benefit: productivity. How much faster, how much leaner, how much more output per head. What those conversations often don’t cover is what AI could be doing to the people who have to use it? And are businesses investing in the wellbeing needs of those expected to drive it?
The evidence suggests they are not – and the gap is widening.
A January 2026 survey by AI consulting firm Section outlined the scale of the disconnect: 74% of the C-suite described themselves as "excited" about AI, while 68% of individual contributors said they felt "anxious or overwhelmed." These are figures from the same organisations, looking at the same technology and experiencing it in entirely different ways. Senior leaders are racing to implement while often the people doing the implementing are struggling to keep up. Closing that distance is one of the more pressing organisational challenges of the moment.
The brain under pressure
The World Economic Forum's 2026 report, The Human Advantage: Stronger Brains in the Age of AI, produced in collaboration with the McKinsey Health Institute, makes the case that brain health – optimal cognitive functioning supported by mental health, sleep, stress management and social connection – is the foundation on which performance in an AI-era workplace depends. The report argues that companies failing to invest in these conditions risk eroding the judgment, creativity and adaptability that determine how well people work with AI
Research from Boston Consulting Group, published in the 2026 Harvard Business Review, goes further. Workers who frequently use AI experience a measurable increase in mental fatigue – a phenomenon the researchers termed "AI brain fry." Employees managing three or more AI agents in their workflow were significantly more likely to experience mental fog than those using one or two. Tasks requiring high levels of AI oversight demanded 14% more mental effort and caused a 12% increase in mental fatigue.
A finding from Workday's January 2026 research adds another layer to this picture. While 85% of employees said AI saved them between one and seven hours each week, they lost 40% of those gains correcting, rewriting or fact-checking AI output. Efficiency is being clawed back by the tool designed to generate it and the effort of that falls almost entirely on the worker.
A divide that is already hardening
The pressure is not evenly distributed. Writing in the Guardian, Professor Nazrul Islam of the University of East London says that AI is creating a new and largely unexamined divide across the workforce – between those who use AI to extend their skills and those whose working lives are increasingly shaped by AI-powered systems of surveillance and control. For higher-autonomy workers, AI can function as a genuine copilot, supporting judgment and clearing space for more complex thinking. For many others, it functions as a manager: setting pace, measuring output and flagging underperformance through systems that workers have no visibility of.
A third of UK employers are already using "bossware" technology to monitor workers' online activity, according to recent data cited by Islam. As algorithmic management spreads from warehouses and logistics into professional and knowledge work environments, the wellbeing implications will follow. Amazon's software engineers have reported being monitored and pushed to use AI to hit productivity targets, even when using it makes their work slower.
More than one in five employees globally already experience symptoms of burnout. Proactive investment in employee health – including brain health – could increase global GDP by up to 12%. One organisation cited in the WEF report whose wellbeing programme combined coaching, internal workshops and interpersonal skills training for managers delivered an 11.6x return on investment.
The governance gap
The infrastructure to manage any of this is largely absent. The Section survey found that only 27% of individual contributors had received company AI training and just 32% reported clear access to AI tools. Dennis Stolle, head of applied psychology at the American Psychological Association, commented on the report: "We never designed the workplace for these kinds of tools. We've just foisted it on people."
The research is direct about where responsibility sits. AI-enabled changes must be paired with deliberate investment in mental health skill support – resilience, adaptability, self-regulation – and without this pairing, organisations face the risks of declining employee wellbeing on one side and diminishing returns from AI investment on the other.
Workplace wellbeing has absorbed considerable pressure in recent years. The pandemic reshaped how and where people work, while the cost-of-living crisis hit financial security and mental health simultaneously. AI adds a new and less understood dimension to that challenge – one that is arriving faster than most organisations have governance structures to manage, and one that will require senior leadership, not just HR teams, to take seriously.
