The rapid spread of artificial intelligence in organizations has produced an equally rapid regulatory response, but one that is far from uniform. Three major geopolitical blocs are taking different paths, and companies operating globally must contend with all three.
The United States bets on deregulation
On January 23, 2025, President Trump signed executive order 14179, aiming to remove obstacles to American leadership in AI. The text explicitly calls for the development and use of artificial intelligence to occur without bias, including ideological bias. A few days later, on January 31, a second executive order reinforced the direction: fewer rules, more innovation.
The federal government's message is clear. But the federal regulatory vacuum does not mean an absence of rules: California, Colorado and Utah have already passed their own AI laws, and Texas is considering its own legislative proposals. Companies operating in the American market must therefore monitor both the federal level and the state level, which is moving independently and not always consistently.
Europe enforces the AI Act
The European Union has chosen the opposite approach. The AI Act imposes precise obligations on organizations that develop or use artificial intelligence systems on European territory, with a risk classification system that determines compliance requirements. Companies operating in Europe are required to build concrete AI governance and risk mitigation strategies, not merely declarative ones.
China regulates with its own approach
Chinese AI laws follow a logic different from both the American and European approaches. Like the AI Act, they require global businesses to develop robust governance practices. But the specific content and areas of focus reflect national political and industrial priorities that do not always align with Western standards.
What changes in practice for companies
The emerging picture is one of a fragmented global regulatory landscape, where the same technologies are subject to different obligations depending on the country in which they are developed or used. Legal and compliance leaders in international organizations must build the foundations of AI governance today, strong enough to withstand future regulatory changes as well. Waiting for the landscape to stabilize is not a strategy: the landscape will continue to evolve, and those who do not already have a foundation risk falling into a structural disadvantage.
