McKinsey’s 2026 State of AI report reveals a striking gap between AI adoption and financial performance. But I believe an even more fundamental question deserves attention: are we using AI to improve today’s businesses, or to prepare for tomorrow’s reality?
There is a risk in the current enthusiasm around artificial intelligence that receives surprisingly little attention.
Organisations are investing heavily in AI. Employees are becoming more productive. Processes are being automated, software is becoming easier to develop, and intelligent agents are beginning to perform work that previously required substantial human effort.
Yet an organisation can become considerably more efficient while its business model becomes progressively less relevant.
We may be getting much better at doing things that, in a few years, no longer need to be done in the same way—or by the same organisations.
This is the strategic challenge I believe deserves a much more prominent place in boardroom discussions.
The productivity paradox
McKinsey & Company’s State of AI in 2026: On the Road to ROI, published in August 2026, offers a revealing picture of where organisations stand.
Based on responses from 1,719 participants across 97 countries, the study finds that:
- 80% report that AI has improved their personal productivity.
- Only 37% report a positive contribution from AI to their organisation’s EBIT.
- Just 6% qualify as AI high performers, reporting significant value and an EBIT contribution of at least 5%.
- 32% say their organisation has decided against purchasing at least one software product or feature because AI coding tools enabled them to develop the functionality internally.
These are self-reported survey findings, not independently measured financial outcomes. Nevertheless, the pattern is compelling.
AI capabilities are advancing rapidly. Adoption is expanding. Investment continues to grow. But translating individual productivity into structural business value is proving considerably more difficult.
McKinsey identifies an important difference between the organisations achieving substantial value and the rest: nearly three-quarters of AI high performers report fundamentally redesigning workflows, compared with approximately one-quarter of other organisations.
That is a meaningful distinction.
But I believe we need to take the reasoning one step further.
There are three levels of AI transformation
When I look at the strategic implications of AI, I distinguish three levels of transformation.
1. Making the existing business more efficient
This is where most organisations naturally begin.
Employees use AI to write, analyse, develop software, process documents and support decisions. Existing processes become faster and potentially cheaper.
There is genuine value here. Improving productivity, reducing administrative burdens and strengthening service quality are worthwhile objectives.
But the underlying business remains largely unchanged.
The organisation is essentially doing what it already did, only more efficiently.
2. Fundamentally redesigning how the business operates
The second level requires a different mindset.
Rather than asking how AI can support existing processes, leaders reconsider why those processes exist in their current form.
Why does a customer request pass through five departments? Why are decisions distributed across so many systems and management layers? Why does an employee need to retrieve information manually when an intelligent system could assemble it within seconds?
At this level, AI becomes an enabler of a fundamentally different operating model.
Human expertise, automated decision-making, intelligent agents and information systems can be organised around outcomes rather than inherited organisational structures.
This is where substantial improvements in cost, speed, scalability and customer experience become possible.
It is also where transformation becomes difficult. Processes, responsibilities, governance, partnerships and established ways of working all come into question.
McKinsey’s findings suggest that high-performing organisations are considerably more willing to embrace these changes.
3. Reconsidering the relevance of the business itself
This third level is, in my view, the most important—and potentially the most underestimated.
What happens when the assumptions underlying your strategy, business model and competitive position fundamentally change?
Many businesses have evolved in a world where expertise was expensive, information fragmented, coordination difficult and developing software required substantial investment.
These constraints shaped entire industries.
They influenced which services could be offered profitably, how companies organised themselves, which intermediaries were necessary, and where competitive advantages emerged.
AI is beginning to change these economic foundations.
When sophisticated analysis becomes inexpensive, when software can be developed much faster, and when intelligent agents can coordinate activities across organisational boundaries, the economics of entire value chains may change.
Consider financial services.
If personalised financial guidance becomes dramatically cheaper to deliver, the opportunity is not merely to make existing advisers more productive. It may become possible to serve customers continuously, at moments when traditional business models cannot economically justify human involvement.
That raises questions about customer relationships, distribution, pricing, propositions and who ultimately captures the value.
The same applies to insurance, professional services, technology businesses and numerous other industries.
Some existing activities may become commodities. New services may become economically feasible. Partnerships and distribution channels may change. Competitive advantages that were built over decades may weaken, while entirely new ones emerge.
The question is no longer simply how to improve the current business.
It becomes whether we would design the same business if we were starting again today.
When intelligence becomes abundant, what remains valuable?
This is one of the questions I find most interesting.
Historically, organisations have built their positions around scarce capabilities: specialised knowledge, analytical capacity, access to information, technical expertise and the ability to coordinate complex activities.
As intelligence becomes increasingly accessible, scarcity begins to shift.
Trust may become more important than access to information. The ability to act responsibly may become more valuable than the ability to produce recommendations. Customer relationships, unique data, capital, distribution and effective execution may become more defensible sources of competitive advantage.
Not every traditional advantage disappears. Regulation, physical infrastructure, institutional credibility and human relationships still matter enormously.
But the relative importance of these assets may change.
And if what creates value changes, strategy needs to change with it.
This has consequences far beyond AI investment decisions. It influences which capabilities organisations should develop, which activities they should outsource, which partnerships they should pursue and where they should allocate capital.
McKinsey’s finding that nearly one-third of respondents report substituting internally developed software for purchases offers an early illustration. Even established build-versus-buy decisions are being reconsidered.
The implications for business models and industry structures could be much larger.
The danger of strategic inertia
I understand why many leadership teams focus initially on productivity.
Efficiency improvements are tangible. AI projects can be assigned to teams, budgets can be allocated, and progress can be reported.
Questioning the foundations of a business is more uncomfortable.
It may challenge successful propositions, established partnerships, organisational structures and investment decisions made only a few years ago.
There is also a natural tendency to extrapolate from the present. Strategies are often developed by looking at historical performance, current market positions and foreseeable competitive developments.
That approach becomes less reliable when the underlying economics are changing rapidly.
And AI is not happening in isolation.
Organisations are simultaneously navigating geopolitical uncertainty, demographic shifts, regulatory developments, changing customer expectations and broader technological disruption.
The challenge is not to predict every development correctly. Nobody can.
It is to understand which assumptions about the future are becoming fragile, which strategic options deserve attention, and which decisions should be made before circumstances force them upon us.
In turbulent times, the greatest strategic risk may not be making the wrong choice. It may be continuing to execute yesterday’s choices exceptionally well.
Four questions I would put on the boardroom agenda
For leadership teams considering the next phase of their strategy, I would start with four questions.
What becomes abundant?
Which activities, capabilities or services currently command a premium because they are difficult, expensive or time-consuming—and what happens when that changes?
What remains scarce?
Which assets and capabilities will truly differentiate our organisation when intelligence and software become more accessible?
Where will value move?
How might customer expectations, industry boundaries, distribution channels and competitive positions evolve as technology changes the economics of our market?
What would we do differently if we were starting today?
Would we offer the same propositions, employ the same operating model, develop the same capabilities and rely on the same partners?
These questions are not intended to produce radical change for its own sake.
Quite the opposite.
A thoughtful strategic reassessment may confirm that important parts of the existing business remain highly valuable. It may identify opportunities to strengthen competitive advantages that competitors have overlooked.
But it should also reveal where established assumptions no longer deserve unquestioned confidence.
The outcome should be a clearer set of strategic choices, not another collection of disconnected innovation initiatives.
From technological transformation to strategic renewal
McKinsey’s report makes a convincing case that organisations must move beyond AI adoption towards fundamental workflow redesign.
I agree.
But I would extend that argument.
Redesigning how a business operates is not enough if we have not reconsidered what that business should become.
The starting point should therefore not be an AI roadmap. It should be a clear perspective on how technology and other structural developments may reshape the environment in which the organisation creates value.
From there, leadership can evaluate alternative business models, future operating models, investment priorities, partnerships and the organisational changes required to execute.
This does not necessarily call for a lengthy strategy exercise. In rapidly changing environments, strategic direction and practical experimentation should reinforce one another. Leaders need sufficient clarity to act, combined with the flexibility to adjust as new evidence emerges.
Throughout my career, working at the intersection of business strategy, technology and organisational transformation, I have become increasingly convinced that meaningful progress depends on connecting these perspectives.
Technology alone is rarely the answer. Neither is a strategy that cannot be translated into operational reality.
What matters is understanding how economics, business models, customers, technology, people, governance and execution fit together—and making deliberate choices about the future.
That integrated perspective is also the foundation of my work through Omniscend.
The question is not whether to embrace AI
By now, the importance of AI is difficult to ignore.
The more consequential question is what we choose to do with it.
Will we use these capabilities primarily to strengthen the efficiency of today’s organisations?
Or will we also use this moment to reconsider where we create value, how we compete and what role we want to play in the future?
Both matter. But they require different conversations, different decisions and sometimes a willingness to challenge what has made us successful.
McKinsey’s 2026 research is a valuable reminder that adopting technology and creating business value are not the same thing.
I believe the larger lesson is about strategic relevance.
The organisations that thrive will not necessarily be those that make the greatest use of AI. They will be those that best understand how the world is changing—and are prepared to evolve accordingly.
That is a conversation worth having before the market makes the decisions for us.
Roel Moradi is the founder of Omniscend, an independent strategic advisory practice supporting leaders with consequential choices, transformation and technological disruption.
Facing a strategic question about what these developments mean for your organisation? Start a confidential conversation.
Research reference: McKinsey & Company (25 August 2026), The State of AI in 2026: On the Road to ROI. The research findings are McKinsey’s; the strategic interpretations and perspectives in this article are my own.