There is no single model to follow
By Rogério Melfi, IA4FIN ambassador.
Discussion of digital platform regulation and, more recently, artificial intelligence regulation is usually organized around three established models: the American model, oriented toward markets and self-regulation; the European model, structured around rights and principles; and the Chinese model, marked by state control and integration between regulation and industrial policy. Although this framework is useful for understanding the topic’s recent evolution, it is insufficient to explain what is happening in a growing group of countries that do not fully identify with any of these approaches.
What we observe is the emergence of new architects of the digital world: markets under construction operating with significant structural constraints, while also facing a more immediate need to turn technology into economic development. In these contexts, regulation stops being understood merely as an instrument for containing risks and takes an active role in organizing capabilities, encouraging investment and building technological infrastructure.
In the United States, the legal tradition favored innovation and market expansion, with strong protection for intermediaries and regulation fragmented among different agencies. In the European Union, meanwhile, a comprehensive, principles-based approach consolidated, emphasizing fundamental rights protection and regulatory standardization, extending its effects beyond European borders. In China, regulation forms part of a broader logic of state planning, in which digital platforms are simultaneously drivers of economic growth and objects of political and strategic discipline.
These models respond to specific institutional trajectories and therefore cannot simply be transplanted into other contexts. For countries developing digitally, the central question is not choosing between market freedom, rights protection or state control, but finding a balance between innovation, sovereignty and competitiveness in an environment marked by capacity limitations and structural asymmetries. This balance appears in recurring dilemmas, such as the need to stimulate digital ecosystems without abandoning essential safeguards, or to preserve control over data and infrastructure without compromising integration with global technology and information networks.
In this scenario, recognizing regulation’s dual nature is fundamental. Alongside governance and risk mitigation, there are instruments aimed at promoting inclusion, developing infrastructure, building human capital and creating favorable conditions for innovation. This dimension is particularly relevant in markets still consolidating their digital foundations, where the absence of incentive policies can be as limiting as excessive restrictions.
In this context, many countries have chosen to begin their regulatory journey through frameworks, guidelines and controlled experimentation environments such as regulatory sandboxes, rather than adopting rigid, comprehensive legislation from the outset. This approach allows greater flexibility in response to the speed of technological transformation, reduces the risk of regulatory obsolescence and expands institutional learning capacity over time.
National strategies reflect this diversity of paths. Some countries bet on combining scale with local talent development; others prioritize productivity gains or investment in computing infrastructure. In every case, however, regulation appears integrated into a broader economic development and competitive positioning strategy. There is no convergence around a single model, but adaptations guided by each context’s conditions and priorities.
At the same time, a significant risk emerges from uncritically adopting external models. Importing regulatory structures without proper adaptation tends to produce dependencies extending beyond rules themselves to technological infrastructure, operating standards and data flows. In this sense, regulation can stop being an instrument for organizing markets and become a vehicle for reproducing asymmetries, especially in a global environment where technological capabilities remain highly concentrated.
Brazil occupies a particular position in this landscape. On one hand, it has significant institutional capacity and a recent history of regulatory innovation, evidenced by initiatives that achieved scale and international recognition. On the other, it faces challenges related to regulatory complexity and the need to coordinate different regulatory agendas. Here, the risk lies not in the absence of regulation, but in its fragmentation and the difficulty of strategically coordinating its different instruments.
Building effective regulatory policy for artificial intelligence and digital platforms therefore requires more than adopting preexisting models. It requires connecting governance mechanisms, development incentives and strategies for competitive participation in the global environment. International experience indicates that there is no universal path to follow, but a need for different combinations adjusted to each country’s institutional capacity, economic strategy and stage of technological development.
In this context, artificial intelligence regulation stops being a choice among ready-made models and becomes a strategic construction in which each country defines, from its own conditions, the feasible balance between innovation, protection and competitiveness, recognizing that this balance is not static but evolves as national capabilities change.
📌 Originally published in IA4FIN