AI Is Exposing the Biggest Blind Spot in How We Manage Wellbeing & Performance at work

Despite unprecedented investment in performance measurement, wellbeing initiatives, and people analytics, the reality of work continues to deteriorate:

• Globally, depression and anxiety account for an estimated 12 billion lost working days each year, costing the world economy around one trillion dollars in productivity.
• Employee engagement is at historic lows.
• In Europe alone, work-related mental health issues cost organisations well over one hundred billion euros annually through absenteeism, long-term sick leave, turnover, and reduced performance.

These outcomes persist despite decades of research, experience, and good intentions. They suggest that something fundamental is missing in how we think about performance and wellbeing at work.

What Performance Actually Depends On

For more than thirteen years, my work has focused on wellbeing and performance at work. As a trainer, coach, and advisor, I have worked with teams and leaders across industries, organisational contexts, and countries. Alongside this practical work, I have continuously engaged with the research that underpins this field, including stress and burnout, motivation and engagement, cognitive load, leadership behaviour, psychological safety, and organisational dynamics.

While research on wellbeing and performance is extensive, it is also fragmented. Different disciplines study different aspects of human functioning at work, often in isolation. Each offers valuable insight, but none captures the human being as a complex, interconnected whole. As a result, we have understood parts of the picture, without ever bringing them together in a way that reflects how people actually function at work.

I created Human Performance Intelligence™ because of a very specific realisation. AI-enabled work environments would, for the first time, make it possible to observe human performance as it unfolds, rather than only through its consequences. Through my investment in and collaboration with Kaamfu , I saw that AI could surface patterns of work, interaction, and pressure at a scale and continuity that had not previously been possible. This created a unique opportunity to integrate what we already know about wellbeing and performance into a system-level framework that could finally be applied in real organisational contexts.

Human Performance Intelligence™ is the result of that work. It is not a new theory layered on top of existing knowledge. It is a deliberate synthesis of decades of established research, brought together to explain how human performance is constructed over time, under real conditions, in complex work systems.

The Five Pillars of Human Performance Intelligence™

At the core of Human Performance Intelligence™ are five interdependent pillars. Together, they define the conditions under which human performance can be sustained over time. They do not operate in isolation. They interact continuously as part of a system.

1. Cognitive capacity and load. Human performance is cognitively bounded. Attention, working memory, and executive function impose real limits on how much complexity, information, and decision-making a person can handle without degradation. When cognitive demands exceed these limits, performance becomes unstable. Errors increase, decision quality drops, and mental fatigue accelerates. This pillar asks a simple but often ignored question: can the human mind realistically handle the demands imposed by the work as it is currently structured?
2. Energy, stress, and recovery. Performance is also biologically constrained. Stress responses, fatigue accumulation, and recovery processes determine whether effort can be sustained or whether breakdown becomes inevitable. This pillar treats wellbeing not as an aspiration, but as a hard performance constraint grounded in physiology. Chronic stress, insufficient recovery, and disrupted rhythms do not reduce performance immediately, but they make its collapse predictable. The key question here is whether the system is biologically sustainable under its current intensity and pace.
3. Motivation, meaning, and engagement. Capacity and energy alone do not guarantee performance. People also decide, consciously or not, how much effort they are willing to invest. This pillar explains why performance varies even when cognitive capacity and energy are available. Motivation, meaning, and engagement determine whether effort is applied, withdrawn, or strategically conserved. The central question is whether available capacity will actually be used in this environment.
4. Social and interpersonal dynamics. Performance in organisations is inherently relational. Coordination, learning, and problem-solving depend on trust, communication quality, and psychological safety. Social friction silently increases cognitive and emotional load, undermining performance even when individuals are capable and motivated. This pillar asks whether people can coordinate and learn together without social dynamics degrading performance.
5. Adaptive capacity over time. Modern work is defined by continuous change. The most critical performance capability is not short-term output, but the ability to adapt, recover, and remain stable over time. This pillar integrates all others longitudinally. It examines whether a system learns and strengthens through pressure, or whether it gradually drifts toward instability and breakdown. The key question is whether performance becomes more resilient over time, or more fragile.

When one or more of these conditions deteriorate, performance rarely fails immediately. It erodes quietly. Quality declines. Learning slows. Volatility increases. By the time failure becomes visible, the system has often been operating outside human limits for far longer than anyone realised.

A Once-in-a-Generation Opportunity

What genuinely excites me about AI is not its promise of efficiency or automation. It is the opportunity to finally close a gap that has existed in organisations for decades: the gap between what we know about human wellbeing and performance, and how work is actually designed and managed.

We already know that chronic stress, lack of recovery, poor leadership practices, and unhealthy work environments lead to higher healthcare costs, increased absenteeism, higher turnover, and declining performance over time. This knowledge is well established. Intuitively, it also makes sense. When people are exhausted or overwhelmed, performance suffers.

And yet, wellbeing has remained marginal in many organisations. Not because leaders do not care, but because the link between wellbeing and performance has been difficult to act on. Wellbeing does not produce immediate results. It works through cumulative, often delayed effects. Under short-term performance pressure, this makes it harder to prioritise.

AI changes the dynamic

For the first time, AI-enabled work environments make it possible to observe the relationship between human conditions and performance much earlier and much more clearly. Not as a belief or a long-term promise, but as patterns that become visible over time. This does not turn wellbeing into a quick fix, but it does make its role in performance far more tangible and credible in decision-making contexts.

From my perspective, this is a profound shift. It opens the door to designing work environments and management practices that are explicitly human-centred, not in opposition to performance, but in service of it. If insights about cognitive load, emotional regulation, social dynamics, leadership behaviour, and recovery are integrated into AI-enabled systems from the start, organisations can move beyond reactive management and begin designing for sustainable performance.

This is what makes this moment so compelling for me personally. Through my collaboration with Kaamfu , I have the opportunity to participate in shaping AI-enabled work environments at a stage where foundational choices are still being made. If wellbeing and human performance are integrated now, they become part of the infrastructure of work rather than an afterthought.

If they are not, we risk missing out on some of the most significant gains AI could offer. Not only in productivity and performance, but in creating work environments where people can genuinely function well over time.

AI does not determine that outcome. How we design and interpret it does.