Moving from AI-enablement to AI-shaped – The importance of intentionally shaping your operating model for AI

We’re seeing organisations across varied industries accelerating their adoption of AI. Agents built and new use cases introduced across functions – all with the aim of improving productivity, speeding up decision-making and remaining competitive.

Beneath this momentum, however, a more complex picture is emerging. In many cases, AI is being deployed faster than organisations are adapting their operating models to support it. Rather than resolving underlying challenges through intentional operating model design and improvement, layering AI over the top can make the fractures more visible and, in some cases, more pronounced.

This often is a result of introducing AI into environments where workflows are fragmented, decision-making is unclear and data is inconsistent. While this can deliver short-term gains, it can also embed existing inefficiencies more deeply as organisations scale. As a result, some AI initiatives struggle to deliver sustained value; not because of limitations in the technology, but because the organisation itself is not set up to use it effectively.

The issue is not that organisations are moving too quickly on AI. In many cases, the challenge is that adoption is happening without enough attention to the design choices that determine whether value can be sustained.

Organisations need to start addressing the bigger question: ‘What would an operating model designed with AI in mind look like?’

Questions around how work flows, where decisions sit, what should be standardised and which processes are still fit for purpose are often left unresolved. In previous waves of ‘traditional’ transformation, these were central to process redesign and automation. In the current AI cycle, they are sometimes being deferred rather than addressed.

What makes this shift different is the scale and pace at which AI is influencing work across the organisation. Unlike previous technology transformations, AI is not confined to a single function, process, or capability. It has the potential to influence decision-making, operational workflows, customer interaction and knowledge-based work simultaneously.

This breadth of impact makes it harder to treat AI as a standalone technology initiative. The organisations seeing the greatest value are approaching it as an operating model challenge as much as a technology one. To realise sustained value of AI, organisations need to move beyond the question of how to embed AI into existing structures, and instead start to consider what an operating model designed with AI in mind would look like.

What changes in an operating model designed with AI in mind?

Designing an AI-shaped operating model changes the emphasis of how organisations are structured and run. Five shifts matter the most.

1. Work is designed around flows rather than just functions

Functions and roles remain important, but they are no longer sufficient on their own. Greater attention needs to be paid to how work moves across the organisation end-to-end, particularly where AI is being introduced into customer journeys, operational processes, or decision-making. Without this view, AI can improve isolated tasks within discrete functions while leaving broader inefficiencies in place.

2. Decisions become a more important design consideration

As AI supports or automates more activities, greater clarity is required on where decisions sit, how they are governed and where human judgement remains necessary. This is particularly important in areas involving ambiguity, accountability, or regulatory scrutiny. The question becomes centred on how decisions should be structured so that AI can be used responsibly and effectively, rather than simply what AI can do.

3. Operating logic must be explicit

Many organisations still rely on informal workarounds, tacit knowledge and locally understood ways of getting things done. That can be workable in a largely human operating environment, but it becomes more limiting when AI is introduced at scale. If processes, rules and handoffs are not sufficiently clear, AI is more likely to reinforce inconsistency than reduce it. Whereas an AI-shaped operating model demands greater consistency in how work is defined and executed.

4. The operating model functions as a connected system

As AI becomes more embedded, the operating model increasingly needs to coordinate people, process, data, technology and AI-enabled capabilities as part of a coherent whole. Treating AI initiatives purely as technology deployments, rather than operating model changes, risks creating fragmented capabilities that struggle to scale consistently across the organisation.

5. Roles and identities shift

Designing an operating model for AI changes how people experience work, where expertise sits and what individuals are accountable for. As AI becomes embedded into day-to-day activity, roles are likely to evolve beyond task execution towards judgement, oversight, interpretation and exception handling. In many cases, individuals will increasingly work alongside AI-enabled capabilities rather than simply use them as tools.

This creates organisational challenges that are often underestimated. Redesigning workflows is relatively straightforward compared to redesigning how people understand their role, how expertise is developed and what good performance looks like in an AI-enabled environment.

Operating model redesign needs to account not only for process and technology change, but also for the human transition that sits alongside it. Without this, adoption risks becoming inconsistent, fragmented, or overly dependent on individual teams and behaviours.

Where should leaders start?

For leaders, the immediate priority is not just to expand and explore the number of AI use cases, but to also ensure the organisation is equipped to realise value from them over time. This could simply mean simplifying core processes, clarifying ownership of key decisions and making operating logic more

consistent. However, it also means being more deliberate about where AI is applied, ensuring it is intentionally aligned to areas of meaningful value rather than layered onto existing complexity.

As AI becomes more embedded into day-to-day operations, the organisations seeing the greatest value are unlikely to be those deploying the highest number of use cases. They will be the ones that redesign how work flows, how decisions are made and how people interact with AI-enabled capabilities across the organisation. AI may expose the fractures in an operating model. Whether organisations address them intentionally, or scale them unintentionally, is increasingly becoming a leadership question.