I have been closely following the transformation brought about by artificial intelligence in the business world. At the center of this revolution, the role of the CIO has evolved rapidly. It is no longer enough to enable technology. It is necessary to lead the change. And this is where the difference lies between an operational CIO and a truly transformative CIO.
The CIO who acts only as a technical enabler of AI loses the most important part of the equation: the impact on the business. Of course, information security, data architecture, and compliance are fundamental topics, but not sufficient. True transformation occurs when AI is thought to change the way the company operates, and this requires a deep understanding of the business model.
Today, much of the value of generative AI lies in orchestrating multi-agent solutions capable of automating processes, making real-time decisions, and changing the way entire areas work. For this, the CIO needs to go beyond IT. It needs to master strategic design, user experience, service journey. Only then is it possible to align technology with purpose and impact.
Such alignment is still a barrier for many. According to the Gartner CIO Agenda 2025 study, 72% of CIOs worldwide state that artificial intelligence is among the strategic priorities of the technology area. However, only 24% can prove they are generating tangible value with initiatives. This highlights a gap between intent and execution, reinforcing the need for a more active and strategic role of the CIO in the AI journey.
Three key competencies to move out of the lab
If you are a CIO and are still stuck in the experimentation phase, my suggestion is clear: develop three key competencies to turn the tide and deliver real value.
- Strategic and Service Design: Understanding how workflows and experiences connect is essential to building AI solutions that make sense within the business.
- Agile Experimentation: Nothing replaces the ability to test fast, fail fast, and learn even faster. Models like Scrum, Lean, and Design Sprint are great allies.
- Adaptability: AI changes every day. New models emerge, APIs transform, regulations appear. The CIO and their team need to be prepared to rebuild whenever necessary. This is part of the game.
Indeed, a recent study by MIT Sloan Management Review in partnership with BCG indicates that only 11% of the analyzed companies have managed to achieve a positive financial return with AI. What do they have in common? A strong integration between technology and business strategy, as well as clear governance and a focus on value from the outset.
How I have been applying this in practice
At the company where I work as CIO, we decided to democratize access to AI from the beginning. We built an internal platform, a true AI hub, that connects different models (including the main market LLMs) in a single interface, accessible to all 900 employees.
This measure avoids two common mistakes: the uncontrolled use of public tools (which can compromise sensitive data) and the limitation of AI use to isolated niches. Here, everyone has access, from customer service to leadership.
Additionally, we created a public innovation roadmap, updated twice a week, that clearly shows the ongoing projects, their phases, deliveries, and next steps. This generates transparency, engagement, and accountability.
Another front is the monthly AI workshops, covering topics such as autonomous agents, prompt engineering, LLM comparisons, among others. More than 400 people actively participate. And most importantly: we have a C-Levels council that prioritizes AI initiatives based on business returns.
This type of structure and initiative is increasingly present in Brazil. The IDC Latin America AI Spending Guide 2025 estimates that Brazilian companies are expected to invest over $1.9 billion in artificial intelligence solutions this year. The main focuses are process automation, customer service, data analysis, and decision support. In other words, the local market already sees AI as a strategic pillar, no longer as an isolated experiment.
AI is no longer a laboratory — it is a value platform
If I could give advice to other CIOs, it would be: stop treating AI as a laboratory experiment. Choose small use cases with high potential impact and quick implementation, and put them into production. Even if imperfect, these field tests will provide valuable feedback to improve the solution.
True progress happens when the development team and end users work together. Continuous collaboration between technology and business generates more relevant, effective, and lasting solutions.
In the end, good AI is AI that works in the real world. The CIO who understands this, who builds together with users, stops being just a technology manager to become a business transformation protagonist.
