UNESCO guidelines for use of AI systems in courts and tribunals – draft published

UNESCO, as part of its AI and the Rule of Law programme, has published draft guidelines for the use of AI systems in courts and tribunals.  

The Guidelines aim to offer comprehensive guidance to courts and tribunals to ensure that the deployment of AI technologies aligns with the fundamental principles of justice, human rights, and the rule of law.

The draft Guidelines are open for consultation, so further development can be expected.  They follow UNESCO's survey and research that use of AI by courts and tribunals is growing.  The Guidance sets out reported use cases globally of AI by courts, judges and lawyers. These include:

• language translation (Supreme Court of India), natural language processing to predict judicial decisions from the Court of Justice of the European Union, generating summaries of judicial decisions to improve accessibility (Argentina). 
• UNESCO notes issues, too - cases of judges and lawyers issuing decisions or submitting legal documents that included references to non-existent rulings due to the use of AI have been reported in the US, South Africa, and Brazil.

A few official principles, rules or guidelines on how AI can be used ethically and responsibly in the administration of justice do exist, such as judicial guidance in the UK (see our summary here) and the note by the Council of Europe's working group for the efficiency of justice (see our summary here).  However, formal guidance on adequately using these tools in the justice sector is otherwise limited.  UNESCO say that that the need for guidance is increased due to new AI laws in place (e.g. the EU AI Act) or on the horizon (e.g. in the UK), and risk that AI involved in judicial decisions may have cascading effects on the whole judicial system (e.g. precedent decisions, or case management).

The draft Guidelines set out thirteen principles for organisations and individuals in the judiciary when adopting and using any type of AI system, including:

1. Protection of human rights
2. Proportionality
3. Safety
4. Information security
5. Awareness and informed use
6. Transparent use
7. Accountability and auditability
8. Explainability
9. Accuracy and reliability
10. Human oversight
11. Human-centric design
12. Responsibility
13. Multi-stake governance and collaboration 

Many of the above principles are explained further in UNESCO's 2022 Recommendation on the Ethics of Artificial Intelligence.

The Guidance goes on to provide specific guidance for organisations and individuals who are part of the judiciary.

The guidelines are open for public consultation until 5 September 2024.

If you would like to discuss how current or future regulations impact what you do with AI, please contact Tom Whittaker, Brian Wong, Lucy Pegler, David Varney, or Martin Cook. For the latest on AI law and regulation, see our blog and newsletter.