Since August 2024, European public administrations have stopped viewing artificial intelligence as merely a technical tool. The entry into force of the European Artificial Intelligence Regulation, known as the AI Act, has significantly raised the standards for using systems capable of influencing public decisions. Although the application of the regulation will be gradual, stricter obligations will come into effect starting August 2, 2026. This timeline should not be seen as an invitation to wait, but rather as a limited opportunity to review current uses, define responsibilities, and strengthen internal control mechanisms before compliance becomes mandatory. The AI Act is not just another technological regulation. It requires moving beyond a purely functional approach to artificial intelligence and recognizing its legal impact. It establishes prior controls, demands human oversight, and reinforces accountability, directly affecting the design, procurement, and use of these solutions in the public sector. It is no longer sufficient to comply with the law in a formal manner or to react to problems; it is now essential to demonstrate that the use of artificial intelligence has been adequately evaluated and controlled. Spain has quickly advanced in creating governance structures and institutional designs to implement the Regulation. The real challenge lies in each administration's ability to integrate artificial intelligence into its daily operations. Key questions arise: identifying the tools used, establishing clear internal policies, training staff, and ensuring effective human oversight are no longer recommendations but minimum requirements. However, in many public agencies, these aspects are still underdeveloped. This gap between regulation and practical application of artificial intelligence is evident throughout the European Union. Effective compliance with the AI Act will depend on both the new rules and the actual capacity of administrations to transform routines, processes, and organizational culture. In this context, Cantabria presents an interesting case. Although it does not play a prominent role in the state governance of the AI Act, its manageable administrative structure and lower fragmentation facilitate the implementation of responsible AI control and usage policies. This advantage is particularly relevant in high-impact sectors, such as public health, where digitalization and the use of artificial intelligence are already a reality in management and decision-making. In this area, human oversight and traceability move from theoretical concepts to immediate operational challenges. Therefore, Cantabria can serve as an effective testing ground, not for regulatory leadership, but for its operational capacity. In a smaller territory, it is more feasible to identify uses, establish common criteria, and reinforce legal control than in larger and more complex administrations. The challenge is not whether artificial intelligence will be incorporated into the Administration, as this is already happening, but whether administrations will be able to regulate its use in a timely manner, with clear legal criteria and effective controls. Cantabria already has initiatives underway that provide a solid foundation. The question is whether it will be able to take advantage of this window before compliance shifts from being a recommendation to an unavoidable obligation.