The issue of artificial intelligence in the workplace, as has been the case with many technologies in the past, is almost always approached through a lens that seems self-evident. We talk about maturity as stages to be completed, as if organizations were following a relatively orderly trajectory, moving from embryonic uses to a more advanced and controlled state. This way of framing the issue has the advantage of simplicity because it gives the impression that the difficulties observed are temporary, whether in terms of skills, tools, or methods, and that all that is needed is to continue the effort for the situation to be resolved.
However, as we observe these practices, this interpretation becomes less and less convincing, not because businesses are incapable of progressing, but because what we interpret as a lack of maturity often corresponds to something much more structural. In many cases, organizations are not stuck because they don’t know how to do something, but because these practices are beginning to have effects that go beyond the technical realm, influencing decision-making and revealing trade-offs that were previously informal and went unnoticed.
The notion of maturity then allows us to reformulate this blockage as a problem of timing or preparation, when in reality it is a limitation that the organization has chosen not to address. It makes the idea of simply waiting acceptable and allows us to avoid making a decision. But the closer we look at these situations, the harder it becomes to believe that the problem will be solved by accumulating more tools or change management experts.
In short:
- The concept of AI maturity is appealing because it avoids confronting the reality of how businesses operate.
- It allows complex choices to be presented as technical steps or acceptable delays.
- This concept ultimately obscures the real issue: the implicit trade-offs about what the company chooses to transform and what it chooses not to.
- An organization’s position on AI depends more on its willingness to reorganize than on its level of maturity.
- The transition to a “AI First” business is primarily an organizational issue, not a technological one.
An accumulation that leads nowhere
Maturity models reinforce this misunderstanding by suggesting that gradual progress will automatically lead to scaling (The problems that arise when AI scales up): more data, more methods, more specialized teams, and AI will eventually establish itself in operations. But experience shows that this accumulation can remain largely ineffective once AI begins to touch on areas where the question is no longer whether it works, but who decides, who takes responsibility, and according to what rules.
Today, we see organizations that are well-equipped, sometimes even highly competent, yet remain confined to limited uses, not because of a lack of competence, but because AI, as soon as it ceases to be a subject of discovery or experimentation, comes into conflict with operating modes that were never designed to absorb it (Prepare the business and work before integrating AI and How can you prepare your organization for AI?). Adding capabilities in this context only makes the gap between what is technically possible and what the organization actually accepts more visible.
Maturity, in this context, becomes an easy excuse to convince oneself that time will do the work, even though there is no indication that the necessary trade-offs will ever be made.
Maturity never exists everywhere
Another problem with this concept is its globalizing nature. We talk about the maturity of a business as if it were a homogeneous whole, whereas the uses of AI are profoundly heterogeneous. The same organization may be very advanced in certain areas, while remaining extremely cautious or even restrictive in others. It may accept AI as long as it optimizes, but then slow it down as soon as it transforms, not because of inconsistency, but because not everything is equally tolerable and therefore negotiable.
Reducing this situation to an average level of maturity amounts to smoothing out differences that explain most of the trajectories observed. It is these internal differences, these areas of acceptance and resistance, that determine what sticks and what disappears. Ignoring them in favor of an abstract scale is to miss the point.
Finally, maturity plays a role in delaying decisions. As long as we are not “mature enough”, it is possible to postpone certain decisions, extend experimentation phases, and observe without making a decision. This time can be useful, but it becomes problematic when it is used to avoid choices that are not part of progressive learning, but rather explicit decisions about what to delegate, what to control, and what to accept being challenged in day-to-day operations (Taking back control of enterprise design: intention before tools and If your business isn’t designed for AI, it will end up being designed by AI).
Through repeated use, the notion of maturity ends up producing a paradoxical effect. It gives the impression of helping to understand the situation, when in fact it masks the nature of the problem. It transforms organizational tensions into technical steps and suggests that time will do the work instead of conscious decision-making. Moving beyond the myth of maturity does not mean denying the differences in capabilities between businesses, but recognizing that the central question is not where we stand on an abstract and theoretical scale, but what we are prepared to accept when AI ceases to be a simple marginal addition.
Bottom line
Maturity in terms of AI remains an appealing concept because it allows us to talk about AI without talking about how the business actually operates. It offers a convenient way to explain delays and transforms difficult choices into technical steps, and sacrifices into simple setbacks.
However, it ultimately obscures the essential issue. What distinguishes organizations in their approach to AI is not their position on a maturity scale, but the way in which they decide, often without saying so, what to transform or challenge (5 archetypes of businesses facing AI). As long as this question remains off the table, maturity will continue to serve as a convenient smokescreen where the issue is less about gradual learning than about conscious organizational design.
This is where the discussion about the path to becoming an AI First company ceases to be a technological debate and becomes a broader business issue.
To answer your questions…
Maturity in AI is attractive because it allows us to talk about AI without questioning how the business actually operates. It transforms complex strategic choices into simple technical steps and makes sacrifices seem like temporary delays. Ultimately, it masks fundamental decisions about what the organization is willing or unwilling to transform. This reassuring framework avoids debate about the necessary structural changes, even though that is where the real issues lie. Practical implication: go beyond maturity to question the real organizational choices.
The difference is not a matter of maturity, but rather the trade-offs that businesses make when it comes to AI. Some are willing to challenge their operating methods, while others seek to preserve the status quo. These decisions are often implicit, but they largely explain the differences observed. Until they are acknowledged, maturity serves as a facade. Practical implication: making these trade-offs explicit is key to moving forward.
By presenting the adoption of AI as a gradual process, maturity transforms resistance to change into issues of timing or skills. It avoids addressing questions of governance, responsibilities, or processes. This shift in the debate makes it possible to postpone difficult decisions without having to make them. Maturity then becomes a smokescreen rather than a management tool. Practical implication: identify what is deliberately left unchanged.
The path to becoming a AI First business quickly goes beyond technology. It involves rethinking how the business makes decisions, operates, and creates value. At this stage, it is no longer a matter of gradual learning, but of conscious organizational design. The obstacles are less about tools than about structural choices. Practical implication: treat AI as a global strategic issue, not as a technical project.
An approach focused on maturity creates the illusion that progress is automatic with time and investment. It depoliticizes decisions that are nevertheless strategic and sometimes intentional. This view prevents us from understanding why some businesses stagnate despite their efforts. Without explicitly questioning the status quo, maturity becomes an excuse. Practical implication: supplement the maturity assessment with an analysis of actual choices.
Image credit: Image generated by artificial intelligence via ChatGPT (OpenAI)
Related articles:
Why the widespread adoption of AI by consumers says nothing about its future in the workplace
Who is handling your artificial intelligence projects? Probably not the right people.
AI First: lots of debate and few answers
What works today with AI without any particular effort
Why AI initiatives always fail for the same reasons
5 archetypes of businesses facing AI
