Our Blog

It’s rare to go a day without encountering in the media a new revolution brought by Artificial Intelligence. New developments, applications, and utilities of AI systems rapidly emerge, promising to radically change our daily lives.

What is undeniable is that the transformative power of AI at all levels leads us to the conviction that we are facing a Revolution destined to change how we work, live, and interact. Therefore, the European Union, through the new Artificial Intelligence Regulation, seeks to safeguard the fundamental rights of individuals, with a firm intention of not falling behind in technological development and thus economic development as well.

AI is not a recent creation; it dates back to the 1950s and, over the years, has gone through periods of growth as well as decline. In 2023, with the introduction of Chat GPT into everyday life as a consumer good accessible to everyone, AI experienced a global explosion.

In light of these circumstances, the European Union stepped up its efforts in developing a regulation, the processing of which had begun in 2021. On December 8, 2023, under the Spanish presidency of the European Union, the Institutions – Parliament, Commission, and Council – reached a historic agreement on the new European AI Regulation, marking a milestone globally due to the regulatory export force that Europe has through what is known as the Brussels Effect. In principle, this effect will turn the approved text in Europe into the global standard in the field, or not.

The agreed text was endorsed by the COREPER, the European Committee of Permanent Representatives, on February 2, so it will see the light of day in a relatively short period of time, entering into force two years after its publication, as a general rule. The European Union has had to balance the protection of security, respect for fundamental rights and the values of the Union, with the no less important interest in promoting investment and innovation within its borders, due to a primary need for survival against great powers such as the United States and China, among others.

The risk of AI, according to the Regulation

The European Artificial Intelligence Regulation classifies AI systems by levels of risk, imposing stricter requirements and rules depending on the level of risk that the use of the system in question implies for Fundamental Rights. AI systems of minimum or no risk will not imply regulatory obligations, although it is recommended that they follow voluntary standards to be more reliable for citizens.

AI systems of limited risk will be subject to transparency obligations (so that users are aware that they are interacting with an AI system), they will have to have human management and supervision, technical robustness and security, respect for privacy and data governance, and respect diversity, not engaging in discrimination or disloyalty. AI systems of high risk will have to comply with obligations both in the design phase, as in the production phase, and in the marketing phase, relating to risk management, data governance, technical documentation, automatic record logs, transparency, human supervision, cybersecurity, and conformity assessment.

Finally, AI systems of unacceptable risk, such as social classification systems, emotional recognition in certain areas, decision manipulation techniques, among others, are prohibited by the Regulation. Beyond the European AI Regulation, there are areas outside the scope of Union law, such as systems created for military or defense purposes, for research or innovation, for non-professional use, and matters of National Security within the exclusive and excluding competence of European States.