AI has the potential to transform every part of the modern value chain. From research and development to manufacturing, supply chain, and commercial operations, its impact is already being felt across the globe. Companies everywhere are looking to these tools to deliver a new era of speed, productivity, and insight.
Yet many leaders are finding that even with advanced tools in place, progress is slower and more fragmented than expected. Teams often find themselves structured in ways that inadvertently limit the technology's potential. When we look at the human side of AI transformation, one shift stands out as foundational: leaders must help their organisations become better collaborators.
When AI meets old silos
Many sectors have long operated through highly specialised functions. Data, systems, and expertise often sit within tightly defined boundaries, shaped by traditional departmental structures and internal risk management. Historically, this structure provided clear lines of accountability and ensured that deep functional expertise was applied to every stage of a project.
However, AI has rapidly changed the requirements for success. AI thrives on connection. Data flows and insights compound when different teams (from discovery and operations to marketing and commercial) learn together. When information remains locked in functional silos, AI’s impact is immediately constrained.
Progress is often limited by very human tendencies: protectionism over data, attachment to functional ownership, and a reluctance to open up work across boundaries. These behaviours can persist even after new technology arrives. As a result, organisations see isolated AI initiatives, duplicated effort, and prototypes that never scale beyond the team that created them. On paper, activity is high. In practice, the value remains fragmented.
The JCB digger analogy
We’ve heard this challenge captured by an analogy that describes how many organisations currently operate:
If you think of a metaphor for how we largely work today, we're all working in a field, digging in our own little areas. At the end of the day, we come together, and we all show each other what we've done, and we move on. All of a sudden, we are no longer using a shovel. We are each sitting in a JCB digger that can excavate the whole field on our own in three minutes. But what nobody's figured out is that it's no good if we all sit in our JCB digger and we excavate the field in three minutes and then we all smash into each other. How are we going to re-organise the way that we work together? To collaborate differently and increase our collective collaboration with the new tools?
To get the most out of AI, there needs to be a move away from this "patch-based" mentality. This requires a fundamental shift in how ownership is viewed. Prioritising the boundaries of individual departments over the flow of information across the business often leads to diminished returns on technological investments. A JCB is designed to reshape the entire landscape, rather than simply deepening an individual trench.
Engineering the conditions for collaboration
Bridging this gap requires a deliberate effort to reward cross-functional outcomes. Leaders can move beyond treating AI as a functional tool and start treating it as a connective tissue for the entire organisation.
This involves addressing the human side directly. It means identifying where protectionism is stalling data flow and where the desire for functional control is preventing a pilot from becoming a company-wide breakthrough. Building a culture of collaboration requires leaders to model a new kind of openness and to actively dismantle the silos that keep people digging in isolation.
When the focus stays on the human conditions required for AI to thrive, organisations can move away from fragmented prototypes and toward true transformation. The technology provides the potential, but the quality of collaboration determines whether teams are just digging deeper holes or building something entirely new.