The Legal Regulation of Artificial Intelligence

The emergence of artificial intelligence (AI) in everyday life has almost completely transformed the legal framework. Law and justice no longer face only the norm and its judicial intricacies, but must also address autonomous decision-making systems, with elements of machine learning and a concerning level of algorithmic opacity. It is precisely in this context that the Spanish and international legal communities must define the limits and establish the necessary safeguards to ensure the proper use of AI, while guaranteeing fundamental rights and legal certainty.

As previously mentioned, AI can be considered a disruptive legal tool, as it is not merely a technology but rather a new subject of legal analysis. This gives rise to numerous questions regarding the allocation of civil and criminal liability, the transparency of algorithmic decision-making processes, and the protection of fundamental rights. The High-Level Expert Group on AI of the European Commission has stated that these systems must be "lawful, ethical and robust." However, a central question that must be addressed is whether the current legal frameworks are equipped to handle the dynamic complexity involved in the use of artificial intelligence.

AI entails an algorithmic opacity in which the inability to audit the reasoning of these complex systems directly collides with the principle of reasoning administrative acts and with the right to a fair trial, as established in Article 6 of the European Convention on Human Rights.

Another issue to consider is that algorithms are not neutral. There is a real risk that they may amplify existing biases, as demonstrated by the COMPAS case in the United States—used to predict criminal recidivism and criticized for its racial bias.

It is in this context that an urgent question arises: Who is responsible for an erroneous decision made by an autonomous AI system?

The European Union has responded swiftly with its regulation: the AI Act, which establishes a legal framework for the control of AI systems. It bans systems deemed to pose an unacceptable risk, imposes strict obligations on high-risk systems (such as those in health, education, or justice), and provides the necessary transparency measures for AI usage.

Even more positively, this regulation is not isolated; it operates in conjunction with other instruments such as the General Data Protection Regulation (GDPR) and the Charter of Fundamental Rights of the European Union.

Looking beyond Europe, the regulation of AI varies globally. In the United States, the approach is less regulatory and more self-governed. In 2022, the White House published a Blueprint for an AI Bill of Rights, which, while not legally binding, sets forth guiding principles on issues such as privacy, algorithmic fairness, and transparency.

In Asia—more specifically in China—the legal framework governing AI prioritizes national security and is firmly controlled by the State. This may reflect the geopolitical value of AI beyond the realm of law. This governmental control is materialized in the Provisions on the Administration of Algorithmic Recommendations.

The integration of AI into the legal domain demands a structural rethinking of the law itself, potentially leading to a general theory of automated algorithmic law. Legal scholarship is already exploring concepts such as algorithmic due process and the need for explainable law.

In conclusion, artificial intelligence should not be seen as a threat, but rather as a stress test for the foundations of our legal systems, which must adapt swiftly and effectively to new realities. For now, AI raises more questions than answers—questions that must be addressed through strong and coherent legislation. Only through proactive and ethical legal action can we ensure that technological development serves human dignity and the rule of law in which we live.