Organisations are investing heavily in artificial intelligence platforms, pilots and employee training, but the biggest challenge in scaling AI may not lie in the technology itself. It may lie in whether people are willing and able to change the way they work.
According to Terragni Consulting, the human behaviour and engagement-focused consulting firm, organisations often treat AI rollout as a technology deployment exercise, with training and successful pilots being viewed as indicators of adoption readiness. However, the transition from a successful pilot to enterprise-wide adoption can expose a less visible challenge — behavioural friction.
“AI transformation is fundamentally behavioural” is the central proposition behind Terragni’s latest AI adoption framework, Vector, which looks at the human factors that can slow down AI adoption inside organisations.
The issue, the firm argues, is that resistance to AI rarely appears as outright rejection. Instead, it can manifest through delayed approvals, pilot fatigue, over-governance, repeated escalation, excessive validation and performative compliance. By the time such resistance becomes operationally visible, adoption may already have slowed.
One of the critical areas identified by Terragni is middle management.
Middle managers play a crucial role in translating organisational strategy into execution, shaping team behaviour, determining how quickly new processes are adopted and creating the environment in which employees experiment with new technologies.
At the same time, AI can potentially challenge the very structures around which middle-management roles have traditionally been built — including authority, expertise, control and relevance.
This creates what Terragni describes as a “middle-management paradox”: the people expected to accelerate AI adoption may themselves experience uncertainty about what AI means for their role and influence.
As a result, resistance may become psychological before becoming operational.
Terragni's framework identifies three broad forms of friction that can influence whether employees embrace new AI-enabled ways of working.
Cognitive friction: Employees may not fully understand how AI changes their role or responsibilities.
Identity friction: Employees may question what happens to their expertise, value or professional identity if AI performs parts of their work faster.
Context friction: Even when employees are willing to change, organisational systems, incentives and workflows may continue rewarding established behaviour.
Together, these factors can create a gap between AI availability and actual AI adoption.
Terragni's approach proposes that organisations should assess behavioural readiness alongside technological readiness.
Its Vector Solution Framework is structured around three stages — Diagnose, Map and Accelerate.
The first stage identifies behavioural barriers such as readiness, managerial resistance, workflow friction, adoption anxiety, role ambiguity and incentive misalignment.
The second maps these issues through tools such as AI readiness heatmaps, behavioural risk dashboards, resistance maps and adoption archetypes.
The final stage prioritises interventions designed to make AI adoption more measurable, actionable and trackable.
The approach is particularly relevant for organisations attempting to move beyond experimentation — whether they are scaling generative AI internally, moving successful pilots into enterprise-wide deployment, redesigning operating models or increasing utilisation of AI tools.
The shift has implications for how organisations measure AI success.
A successful technology deployment does not necessarily mean that employees have changed their behaviour. Similarly, completing an AI training programme does not automatically mean that employees will integrate AI into their everyday workflows.
For businesses, the next phase of AI transformation may therefore require a broader question: not simply whether the organisation has implemented AI, but whether its people, processes and incentives are enabling people to use it.
“Most firms optimise systems. We optimise human adoption,” is the positioning behind Terragni's Vector proposition.
As AI moves deeper into organisational decision-making and everyday workflows, the ability to understand and address human behaviour could become as important to transformation outcomes as the technology itself.