As AI becomes embedded in everyday work, a consistent pattern is emerging across organisations: existing organisation designs are being stretched beyond what they were built to support. This is not a technology problem. It is a design challenge.
Traditional organisation models were optimised for stability, predictability, and human-bounded decision cycles. AI changes all three. It compresses time, redistributes expertise, and alters how information flows across the enterprise. As a result, organisations are being forced to reconsider how authority, coordination, and accountability are structured.
What we see in practice is that organisations are exploring a range of design responses, with varying degrees of success.
One common shift is toward flatter, more horizontal structures. As AI reduces the cost of accessing information and generating insight, the need for multiple layers of approval diminishes. Organisations that adapt well use this opportunity to push decision-making closer to the work, supported by clearer decision rights rather than additional oversight. Those that struggle often retain hierarchical controls while expecting speed — creating bottlenecks instead of momentum.
Another emerging pattern is the rise of networked or modular designs. Instead of organising purely around functions, some organisations are redesigning around flows of value — forming cross-functional teams that assemble and reassemble as priorities shift. AI acts as an enabler here, allowing knowledge to travel across boundaries more easily. However, without explicit coordination mechanisms, these designs can fragment quickly.
We also see experimentation with dual operating models. In these structures, a stable core supports reliability and governance, while more adaptive units are given autonomy to experiment with AI-enabled ways of working. The success of this model depends less on where the boundary is drawn, and more on how clearly it is defined. Ambiguity about which rules apply where often undermines the intent.
A fourth pattern involves redesigning roles rather than structures. As AI takes on portions of analysis, execution, or monitoring, roles are being reshaped around judgement, integration, and decision-making rather than task completion. Organisations that invest in redefining roles early tend to experience smoother transitions than those that simply add AI on top of existing job designs.
Across all of these patterns, one lesson stands out: organisation design choices now carry higher consequences, faster. Misalignment shows up quickly — in duplicated effort, decision paralysis, or over-centralisation. Conversely, well-designed systems create leverage, allowing fewer people to coordinate more effectively.
Future-ready organisations are not converging on a single “best” structure. Instead, they are becoming more deliberate about design — treating structure, decision rights, and coordination mechanisms as evolving assets rather than fixed charts.
The pattern from the work is clear: in the AI age, organisation design is no longer a background consideration. It is a primary leadership discipline — and one that determines whether technology becomes an accelerant or an obstacle.
