Europe has made its move. The European Union has just enacted the Artificial Intelligence Act (AI Regulation), the world’s first comprehensive legal framework in this area. We are entering the era of mandatory algorithmic mitigation. The days of voluntary self-regulation for tech companies operating on the continent are over.
This legal framework will not take effect immediately. Su despliegue será progresivo. Aunque la ley entró en vigor a mediados de 2024, sus disposiciones se escalonarán a lo largo de 36 meses, extendiendo su periodo de adaptación pleno hasta 2027, con excepciones de cumplimiento para ciertos sistemas gubernamentales heredados que alcanzan hasta 2030,.
The regulations are structured according to a strict taxonomy based on risk level. Practices involving unacceptable risk are prohibited. Indiscriminate facial recognition, social scoring (social scoring) and cognitive manipulation are strictly prohibited. Meanwhile, high-risk systems—ranging from biometric screening of résumés to the management of critical infrastructure—will face exhaustive legal scrutiny.
What does this mean for the corporate ecosystem and users?
Companies are undertaking a redesign of their compliance architectures. Los desarrolladores de IA de alto riesgo deberán someter sus modelos a evaluaciones de conformidad de terceros antes de su comercialización,. They will have to certify the quality of training datasets to prevent discriminatory biases and ensure algorithmic traceability through automated logs,. Those who develop generative AI (such as GPT-style foundational models) assume new transparency obligations: they must publish detailed summaries of the copyrighted data used to train their models. To prevent the regulatory burden from stifling innovation in the startups locales, la ley instituye los «espacios controlados de pruebas» (sandboxes (regulatory). These zones will allow companies to test models under real-world conditions without immediate exposure to punitive sanctions.
For citizens, the regulation establishes new enforceable digital rights. Algorithmic transparency is becoming non-negotiable. Users will know when they are interacting with a machine, and they will see clear labels next to altered content or deepfakes and, most importantly, they will have the right to demand a clear explanation when an automated decision affects their health, safety, or fundamental rights.
The Global Regulatory Trilemma: Europe, the United States, and China
Outside Europe, governance varies drastically. The three powers have chosen paths dictated by their political and economic models.
The United States prioritizes technological supremacys. Its approach is decentralized and deliberately flexible. The executive order issued by the White House in late 2023 imposes mandatory guidelines on federal agencies, while leaving the private sector subject to voluntary guidelines. Washington prioritizes entrepreneurial freedom and sustained innovation over legal standardization, fearing that restrictive laws would slow its industrial dominance.
China operates at the opposite end of the political spectrum, but with remarkable regulatory astuteness. Beijing is implementing a “vertical” regulatory approach. Instead of a generic overarching law, it issues specific regulations tailored to each use case: management of recommendation algorithms (2022), deep synthesis (2023), and facial recognition. Its primary goal is to ensure social stability and national security.
One might think that China’s regulatory authoritarianism stifles development. The facts prove otherwise. The success of open-source models like DeepSeek—which is highly competitive and was developed at low cost despite the U.S. embargo on advanced microchips—illustrates the “DeepSeek effect.” Beijing has built restrictive regulation on top of sweeping enabling industrial policies. It has built public supercomputing infrastructure, subsidized corporate innovation, and opened up government databases. State intervention does not stifle innovation; it redirects it toward national self-sufficiency in the face of Western embargoes.
The Outlook for the Future
The global algorithmic ecosystem has fractured. Europe aims to impose the “Brussels effect”. Its strategy is based on the premise that the extraterritorial nature of its regulations will compel any multinational company wishing to operate in its market to adopt EU standards worldwide,.
The challenge facing the European Union through 2028 is a practical one. A human rights protection framework that looks flawless on paper will be of little use if its technical and bureaucratic barriers to entry stifle the continent’s competitiveness. The law has been written. The real test of geopolitical and commercial tensions is just beginning.
Cover image credits: https://www.pexels.com/@bertellifotografia/
Bibliography:
Juárez Aguilar, M. B. (2025). Gobernanza de la inteligencia artificial en China: Regulación que alienta la innovación. Revista de Derecho Económico, 82(2), 3–31. https://doi.org/10.5354/0719-7462.2025.80322
Reglamento (UE) 2024/1689 del Parlamento Europeo y del Consejo, de 13 de junio de 2024, por el que se establecen normas armonizadas en materia de inteligencia artificial (Reglamento de Inteligencia Artificial)[cite: 2]. Diario Oficial de la Unión Europea, L 2024/1689, 12 de julio de 2024 http://data.europa.eu/eli/reg/2024/1689/ojeg/2024/1689/oj
Pérez-Ugena, M. (2024). Análisis comparado de los distintos enfoques regulatorios de la inteligencia artificial en la Unión Europea, EE. UU., China e Iberoamérica. Anuario Iberoamericano de Justicia Constitucional, 28(1), 129-156. doi: https://recyt.fecyt.es/index.php/AIJC/article/view/108247
Source: https://artificialintelligenceact.eu/es/ai-act-explorer/
