The Bottom Line
- Human Judges Remain in Full Control: For businesses engaged in litigation in Spain, judicial decisions will continue to be made by human judges. Artificial Intelligence is strictly designated as a support tool, ensuring that legal outcomes are based on human reasoning and interpretation, not automated processes.
- A Walled Garden for Legal Tech: Judges are only permitted to use AI tools that have been provided and vetted by the justice administration or the judicial council itself. This creates a significant barrier to entry for third-party AI vendors but an opportunity for those who can partner with the government.
- Judicial Responsibility is Non-Negotiable: The new rules reinforce that judges are solely responsible for their rulings. AI-generated drafts require a “complete and critical” human review, meaning legal arguments must still be tailored to persuade a human, not an algorithm.
The Details
Spain’s General Council of the Judiciary (CGPJ) has issued a landmark instruction setting clear rules for the use of Artificial Intelligence by judges and magistrates. This proactive measure aims to create a coherent framework that aligns with new national and European regulations, including the EU AI Act. The primary driver is to manage the risks associated with AI, particularly generative AI, by ensuring its use in the justice system does not infringe on fundamental rights, introduce bias, or undermine the integrity of judicial proceedings. The instruction provides a clear signal that while the judiciary is open to modernization, it will be on its own terms, prioritizing safety and accountability.
The core of the new directive is the principle of effective human control. The instruction explicitly states that AI systems cannot operate autonomously in making judicial decisions, evaluating facts or evidence, or interpreting the law. Judges must maintain constant, conscious, and effective control over any AI tool they use. This is reinforced by the non-substitution principle, which forbids AI from replacing a judge in their jurisdictional duties. Other key principles include upholding judicial independence, ensuring the confidentiality and security of data, and actively preventing algorithmic bias that could lead to discriminatory outcomes.
In practical terms, the rules establish clear boundaries for what is permissible. Judges can use approved AI tools for tasks like legal research, organizing case files, and creating internal summaries or working drafts. However, any AI-generated draft of a judicial resolution must undergo a complete personal review by the judge, who remains exclusively responsible for the final text. The instruction explicitly prohibits using AI for automating or delegating judicial decisions, profiling individuals, predicting behavior, or performing risk assessments. Furthermore, judges are barred from using these tools to process specially protected personal data, ensuring a high level of protection for sensitive information within the judicial system.
SOURCE: General Council of the Judiciary (Spain)
