Executive Summary
Organisations across every sector are moving fast on AI. The technology investments are real, the efficiency gains are measurable, and the competitive pressure to adopt is mounting. But in the focus on systems, tools, and transformation timelines, something is getting missed.
The people are struggling. Not visibly, in most cases. They are still showing up. Still performing. But underneath the surface, the research is unambiguous: AI disruption is generating a wave of anxiety, identity erosion, and disengagement that most organisations are not equipped to address. And at the centre of that problem sits a leadership challenge that rarely gets named directly. Leaders who lack clarity about their own values, priorities, and direction cannot provide it to others. In a period of profound uncertainty, that gap becomes critical.
Left unmanaged, the human cost of AI disruption shows up in attrition, productivity loss, and a workforce that complies with change rather than drives it. This paper looks at what is happening to people inside AI-disrupted workforces, what the evidence says about the downstream effects on business, and how forward-thinking leaders can build a people strategy genuinely fit for what is coming.
01
The Technology Story Is Well Told
The scale of AI's impact on the world of work is no longer speculative. The World Economic Forum's Future of Jobs Report 2025, drawing on insights from over 1,000 global employers representing 14 million workers, projects that 22% of all current roles will either disappear or be fundamentally restructured by 2030. That is 92 million jobs displaced, with 170 million new roles expected to emerge in their place.
The net figure is positive. But the transition is not. Employers surveyed by the WEF acknowledge that 39% of workers' core skills will need to change by 2030, and 77% plan to reskill their workforce in response. At the same time, 41% plan to reduce headcount where AI can automate tasks. Both things are happening simultaneously, inside the same organisations, often without clear communication to the people most affected.
The technology conversation is relatively straightforward. The leadership conversation is harder, and most organisations have barely started having it.
This is where leadership becomes the critical variable. The technology decisions are being made at the top. The communication of those decisions, the culture in which they land, and the confidence people have that there is a considered human strategy behind the technical one, all of that flows from leadership. Organisations whose leaders are operating from a clear sense of purpose and values will navigate this differently from those where leadership is reactive, uncertain, or quietly as anxious as the people they are meant to be guiding.
02
What Is Happening to People
When people read about AI disruption, they tend to process it in relation to their own role. Am I at risk? Is what I do still valued? What happens to my career if this changes faster than I can adapt?
These are not abstract questions. They generate real psychological responses, and the research on those responses is increasingly clear.
A 2025 peer-reviewed study published in Humanities and Social Sciences Communications examined AI adoption across South Korean organisations and found a significant negative relationship between AI implementation and employee psychological safety. As psychological safety declined, rates of depression among employees increased. Critically, the study found that where leadership was ethical and transparent, this effect was substantially reduced. The quality of leadership, not the pace of technology adoption, was the determining factor in whether people came through the transition intact.
Research published in 2025, examining AI-induced job displacement among IT professionals, identified six core psychological themes experienced by people navigating these transitions: emotional shock, erosion of professional identity, chronic anxiety, social withdrawal, maladaptive coping strategies, and a sense of organisational betrayal. These are not unique experiences. They are the documented, sequential stages that people move through when their professional identity is disrupted faster than their capacity to adapt.
The term "algorithmic anxiety" has begun appearing in peer-reviewed literature to describe a specific syndrome: not just the fear of job loss, but a deeper concern about human value, professional purpose, and what meaningful work looks like in an automated future. Unlike previous waves of automation that primarily displaced routine manual labour, AI is reaching into knowledge work, advisory roles, and professions that people built careers around. That changes the psychological stakes significantly.
Leaders who are not themselves operating with clarity and conviction are poorly placed to close that gap.
The PwC Global Workforce Hopes and Fears Survey 2025, covering nearly 50,000 workers across 48 countries, found that only 53% of employees feel strongly optimistic about the future of their roles. Among non-managers, that figure drops to 43%. Workers under financial or career pressure are less trusting, less motivated, and less candid with their employers, creating a feedback loop where organisations cannot get honest signals from their own people at the moment they need them most.
People look up to understand how seriously to take the reassurances they are being given. If the message from the top feels uncertain, rehearsed, or disconnected from a genuine point of view, it does not land as reassurance. It compounds the anxiety it was meant to address.
03
The Business Cost Nobody Is Calculating
Most organisations are modelling the efficiency upside of AI carefully. Few are modelling the human cost with the same rigour.
The downstream effects of workforce anxiety are well documented. Reduced psychological safety correlates with lower performance, less creative problem-solving, and a reluctance to take initiative. Employees who are anxious about their role security become increasingly compliant rather than engaged, focused on self-preservation rather than contribution.
Attrition is the most visible symptom. Major corporations including IBM and Amazon have faced measurable increases in turnover in AI-affected departments during transformation periods. The recruitment and onboarding cost of replacing an experienced professional is significant, but the less visible cost is equally damaging: the institutional knowledge, client relationships, and team capability that leave with them.
There is also a performance paradox embedded in AI-disrupted workforces. Research published in ScienceDirect in 2025, drawing on data from 600 employees across diverse industries, found that AI technostress increases exhaustion and reduces job satisfaction even while it may temporarily boost output. Organisations can see productivity metrics hold steady while the people generating those numbers quietly burn out, making it harder to see the problem coming until it arrives as an attrition event or a team performance cliff.
When leaders are reactive rather than grounded, they send signals that ripple through the organisation faster than any formal communication.
Leadership decisions made under pressure, without clarity, accelerate this. A restructure announced without a clear rationale, a reskilling initiative that feels like window dressing, a town hall that answers no real questions: these are the outcomes of leadership operating without a clear point of view on what the organisation values and where it is headed. The business cost of that is real, and it compounds.
The WEF's research identified skills gaps as the single most significant barrier to business transformation, with 63% of employers citing it as their primary challenge. But the confidence gap and the skills gap are related. People who feel uncertain, undervalued, or anxious about their place in a changing organisation are far less likely to lean into new capability development. And that uncertainty almost always originates in, or is amplified by, what they see in their leaders.
04
What People Actually Need
The research points consistently toward a set of conditions that determine whether people move through disruption constructively or get stuck in it.
Clarity matters more than certainty
People can tolerate a changing landscape if they understand where they stand within it. What erodes psychological safety fastest is ambiguity, particularly when it is paired with the sense that decisions are being made about them without them. The PwC survey found that big payoffs in motivation come when workplaces build trust, offer meaningful work, and provide strategic alignment. These are not perks. They are the conditions under which people perform.
That clarity has to start with leadership. A leader who has not worked through their own relationship to change, uncertainty, and purpose cannot credibly offer it to others. The organisations that communicate most effectively through disruption tend to have leaders who are personally grounded: people who understand their own values, can articulate a genuine point of view, and make decisions that others can read as consistent and considered. That quality of leadership is not innate. It is developed.
Clarity has to start with leadership. A leader who has not worked through their own relationship to change, uncertainty, and purpose cannot credibly offer it to others.
Identity is central to resilience
Much of the anxiety produced by AI disruption is not really about the technology. It is about what the technology implies for who someone is professionally. A financial adviser whose analytical tasks are increasingly automated, or a knowledge worker whose expertise is being challenged by a tool that can replicate it at scale, is not just facing a skills problem. They are facing an identity question. Organisations that address only the former will find the latter resurfaces.
This is as true for leaders as it is for anyone else in the organisation. A senior leader who has built their identity around a particular domain of expertise or decision-making authority is not immune to the identity pressures that AI creates. In some ways they are more exposed, because the expectations on them are higher and the space to show uncertainty is smaller. Leaders who have done the work of understanding who they are beyond their title or function are significantly better positioned to lead through this period.
Connection sustains people through pressure
The research on psychological resilience during organisational change consistently points to the importance of belonging, mentored relationships, and a sense that the people around you are invested in your success. Social withdrawal is one of the documented responses to AI-induced displacement precisely because the disruption severs the relational threads that would otherwise help people navigate it. Leaders who maintain genuine, invested relationships with their teams, who are present and not just visible, provide a stabilising function that no change management program can replicate.
05
The People Strategy of the Future
The organisations that will come through this period with strong, capable, committed teams are not necessarily the ones that move fastest on AI technology. They are the ones that manage the human transition with the same deliberateness they apply to the technical one. And that starts with the people making the decisions.
For leadership teams, a structured development program provides the kind of personalised development that builds genuine clarity. Working across five life domains, leaders develop a considered understanding of their own values, priorities, and the things that drain rather than build their capacity. In a period where strategic decisions have significant human consequences, that self-knowledge is not a soft benefit. It is a prerequisite for making good calls. A leader who understands what drives them, what clouds their judgment, and where their energy actually comes from is better placed to think clearly under pressure, communicate with conviction, and make strategic decisions that hold up over time.
For the broader workforce, the clarity and relational support the research identifies as the decisive factors in navigating disruption well are not delivered by reskilling programs or technology roadshows. They are built through sustained, personalised attention from someone genuinely invested in a participant's growth. That relationship is the mechanism through which people build confidence, process uncertainty, and develop the self-awareness to navigate change rather than be swept along by it.
06
A Different Kind of Readiness
AI readiness is increasingly understood as a technical and operational question. Which processes can be automated? Which roles need to change? Which tools should we adopt and when?
These are the right questions. But they are only half the picture. The other half is leadership readiness: whether the people navigating these decisions have the clarity, the values, and the personal resilience to make them well and bring their organisations with them.
The organisations that treat their people, starting with their leaders, as the primary variable in that equation will be the ones that come through with teams capable of performing at the level the moment demands.
The disruption is real, and it is not slowing down. Strategic clarity at the top creates the conditions for confidence throughout. And confidence, in a period of profound uncertainty, is not a luxury. It is what allows an organisation to move.
References
1.World Economic Forum. Future of Jobs Report 2025.
2.PwC. Global Workforce Hopes and Fears Survey 2025. 49,843 workers, 48 countries.
3.Humanities and Social Sciences Communications. The dark side of artificial intelligence adoption: linking artificial intelligence adoption to employee depression via psychological safety and ethical leadership. 2025.
4.PMC / Taylor & Francis. Psychological impacts of AI-induced job displacement among Indian IT professionals: a Delphi-validated thematic analysis. 2025.
5.ScienceDirect. Insights from the Job Demands-Resources Model: AI's dual impact on employees' work and life well-being. 2025.
6.PMC. Algorithmic anxiety: AI, work, and the evolving psychological contract in digital discourse. 2025.