Table of contents
Three years ago, the most frequently asked question was: “Should we be interested in artificial intelligence?” Today, such doubts are rare. AI has found its way into business strategies, digital transformation plans, and board agendas across virtually every sector of the economy.
Paradoxically, however, as the availability of technology increases, the number of questions does not decrease. On the contrary. The more opportunities the market offers, the more difficult decision-making becomes. In many organizations, the problem is no longer a lack of technology. The problem is determining which technologies are actually worth implementing, what role they should play in the organization, and what specific value they should bring to the business.
After the era of experimentation, it is time to ask about ROI
Over the past few years, many organizations have focused on testing the capabilities of artificial intelligence. Pilots were created, tools were implemented, and transformation programs and employee training were launched. In many cases, the mere fact of using AI was treated as proof of an organization’s innovativeness. Today, the market is in a completely different place.
Chief financial officers and boards are increasingly joining the conversation. They no longer ask whether the company should use AI. They ask what results previous investments have brought, which projects have actually improved efficiency, and whether it is worth allocating further funds to subsequent initiatives. This is a natural stage of market maturation—every technology goes through a moment when the period of fascination ends and the period of accounting for results begins.
Therefore, it is not the number of tools implemented that is becoming increasingly important, but the organization’s ability to demonstrate concrete business value. Reducing process execution time, improving decision quality, increasing team productivity, reducing operating costs, or increasing customer satisfaction are becoming much more important than the mere fact of using AI.
In practice, this also means changing the way technological projects are evaluated. Boards are increasingly expecting not a presentation of a tool’s capabilities, but an answer to the question about return on investment.
AI is not a scarce resource today
Not long ago, organizations that were the first to gain access to new technological solutions built a competitive advantage. Today, the situation is different. Tools using artificial intelligence are widely available. Both global corporations and medium-sized enterprises use similar solutions. This makes the mere presence of AI in an organization no longer a differentiator. Instead, the way it is used is becoming increasingly important. That is why the question “which tool to choose?” often turns out to be less important than the question “what problem do we want to solve?” This seemingly small difference in practice determines the success or failure of many technological initiatives.
PwC data from the “AI Performance Study” (2026) confirms this clearly: 74% of the economic value generated by AI goes to just 20% of companies, and leaders achieve results 7.2 times higher than other enterprises. Importantly, the difference is not the scale of investment, but what AI is used for. Leaders are 2–3 times more likely to be focused on growth and business model reinvention. They are also twice as likely to redesign work processes instead of simply adding more tools.
Organizations that start transformation with technology often quickly encounter difficulties in justifying further investments. It is difficult to demonstrate a return on investment if the business goal, success metrics, and the process to be improved were not defined at the beginning.
The most mature organizations reverse this logic. They first define the business problem, then determine the expected result, and only then choose the technology that can help achieve it.
Most AI-related challenges are not technological in nature
Public debate is dominated by conversations about the capabilities of language models, automation, or new generations of tools. It is much less common to talk about the fact that the biggest barriers to implementation are very often found outside the IT department.
Organizations face questions regarding responsibility for decisions made with AI support, data security, process quality, or the readiness of teams to work in a new environment. In many cases, technology is the least problematic element of the entire undertaking.
It turns out to be much more difficult to determine who is responsible for the change, how to measure its effects, what competencies are needed in the organization, and how to connect new solutions with the company’s existing operating model. This is precisely why—as with any breakthrough technology—a broader view of transformation remains key: not only from the technological side, but above all from the organizational side.
If an organization cannot answer the question of who is responsible for the change, how it will measure the success of the project, and how the new solution will affect the daily work of teams, the problem is not a lack of technology. The problem is a lack of organizational readiness to exploit its potential.
Human + Machine: people and decisions at the center
Many simplifications have been created around artificial intelligence. One of the most popular is the belief that the goal of an organization should be to replace people with technology as much as possible. Meanwhile, the experiences of the most mature organizations show the opposite.
The greatest value is achieved by companies that can combine the capabilities of technology with the experience, knowledge, and competencies of their employees. Artificial intelligence accelerates data analysis, supports processes, and helps identify dependencies. However, humans are still responsible for assessing context, making decisions, and taking responsibility for their consequences. This approach is known as Human + Machine, which we at MINDBOX have been working with for several years. It assumes that effective transformation does not start with technology, but with people, processes, and business goals. Artificial intelligence accelerates analysis, supports processes, and helps identify dependencies. Humans are still responsible for assessing context, making decisions, and taking responsibility for their consequences. Therefore, it is not about choosing between humans and technology. It is about creating a model of cooperation in which both sides mutually strengthen their capabilities.
Our experience shows that this approach works. At MINDBOX, 95% of licensed employees actively use AI tools in their daily work, and the implementation of Microsoft Copilot has resulted in over 6,000 hours saved across the organization. However, this was not a technological project—we started by defining business goals, building an internal Human + Machine competency model, and only then did we implement the tools. We then translate the experience gained internally into projects carried out with clients.
In practice, this approach gives the greatest chance of achieving lasting business results and justifying investment in AI in the long term.
A new competency for leaders
In the AI era, one of the most important competencies for leaders will not be knowledge of specific tools. It will be the ability to make the right decisions in an environment of increasing technological complexity.
Boards do not need to know the details of how language models work or track every new feature appearing on the market. However, they should understand the consequences of technological decisions for the organization, processes, people, and security.
This is precisely why the conversation about artificial intelligence is less and less about the technology itself. Instead, it is increasingly about responsibility, efficiency, return on investment, and building organizations capable of functioning effectively in a new environment.
And this means that the future of AI will depend not on the number of tools implemented, but on the quality of the decisions made around them.
Author
Łukasz Ewertowski
AI Business Transformation & Technology Director, Head of Microsoft Centre of Excellence
