AI and Arbitration: The evolution of AI in the UAE

  • Market Insight 07 March 2024 07 March 2024
  • Middle East

  • Disputes - Technology Risk

This is the sixth article in Clyde & Co’s latest international arbitration series covering Artificial Intelligence (AI) in international arbitration. In this piece, Senior Associate William Prasifka and Knowledge Lawyer Joanna Stewart provide a legal perspective from the UAE.

The United Arab Emirates1 strives to become one of the world’s leading nations in the development of Artificial Intelligence (AI). In 2018, the UAE Government published the UAE National Strategy for Artificial Intelligence 20312 (the AI Strategy 2031), a comprehensive programme aimed at fostering AI innovation.

The AI Strategy 2031 contains eight objectives to establish the UAE as a hub for AI research and development while fostering an ecosystem conducive to innovation. The UAE Office for Artificial Intelligence was also established to enhance AI technology across various sectors.  

As AI technology permeates diverse industries in the UAE, the legal sector is no exception. We predict that AI-related disputes will rise alongside the increased utilisation of AI in legal proceedings.

The UAE Legal System

The legal framework in the UAE is based on both civil and Sharia law principles, with Arabic being the language used in official documents and court proceedings. Arbitration is governed by Federal Law No.6 of 2018 (the UAE Arbitration Law).

Additionally, the UAE encompasses two prominent free zones: the Dubai International Financial Centre (DIFC) and the Abu Dhabi Global Market (ADGM). These zones operate with their own civil and commercial laws and courts, and function as independent arbitral seats, having their own arbitration laws distinct from the UAE Arbitration Law. The DIFC and ADGM Courts operate in English, with a judiciary primarily composed of retired judges from common law jurisdictions. They are the curial courts for arbitrations seated in the DIFC and ADGM respectively. 

The UAE is also home to several separate arbitral institutions, the two most prominent of which are the Dubai International Arbitration Centre (DIAC) in Dubai and the Abu Dhabi International Arbitration Centre (ArbitrateAD) in Abu Dhabi. These institutions can facilitate arbitrations seated in the ADGM, the DIFC, the wider UAE, or foreign seats. 

Regulating Artificial Intelligence 

While the UAE has not issued any laws or guidelines that specifically regulate the use of artificial intelligence in arbitration, several developments have implications for its future use:

  1. Natural Language Processing (NLP) in Arabic is developing at great pace. The Abu Dhabi Technology Innovation Institute (the TII)3  has developed NOOR4, the world's largest Arabic NLP. NOOR can summarise texts and prepare translations. As translation costs in Middle East disputes can represent a major cost and take considerable time in arbitrations, this could become a future efficiency for regional arbitrations.
  2.  Various Arabic large language models are under development. For example, Jais AI has been developed by researchers at Mohamed bin Zayed University of Artificial Intelligence in Abu Dhabi. Such technologies have the potential to provide Arabic language generative AI and technology assisted review (TAR), which could further enhance the efficiency of regional arbitrations.
  3. The Ministry for Artificial Intelligence has issued a white paper on the Metaverse which specifically advocates for transparency regarding the use of any automated or AI driven operations5.  The white paper also highlights that given the interjurisdictional nature of the Metaverse, clear guidelines for dispute resolution will be required. 
  4. The DIFC recently enacted amendments to its Data Protection Law to reference the processing of personal data through autonomous and semi-autonomous systems6. This is the first mention of AI in UAE legislation.  

The DIFC Courts have issued guidance on large language models and generative content generators (GCGs) in dispute resolution. The Practical Guidance Note No. 2 of 2023 on the use of AI in Court Proceedings (the “PGN 2/2023”) came into effect in December 2023, and it sets out the principles and best practice guidelines for parties to uphold, which are: 

  • Accuracy and reliability: any AI-generated content should be verified for accuracy and parties should avoid over reliance. 
  • Early disclosure: the use of AI technology should be disclosed at the earliest opportunity, and before the CMC, along with any potential limitation of the technology or bias. This is in notable contrast to the guidance issued by certain other courts, for example the UK’s AI guidance issued for judicial office holders in December 2023 which does not specifically require disclosure of the use of AI in proceedings.
  • Technology: any technology should be specialist, designed to produce legal content and ensure it is relevant to the DIFC jurisdiction. 
  • Education: Practitioners and clients should be educated on the technology to understand any associated risks, prior to obtaining the client’s consent to use GCGs.
  • Compliance: parties should adhere to the Mandatory Code of Conduct for Practitioners and ensure that the technology used adheres to the relevant laws such as the Data Protection Law 20207 and Intellectual Property Law 20198.

While PGN 2/2023 applies to DIFC court proceedings, similar guidelines may be adapted for the application to arbitral proceedings. 

AI in Arbitration

Given the above developments, it is only a matter of time before AI is widely used in arbitral proceedings in the UAE and wider Middle East. Potential uses include:

Document Review: The benefits of adopting AI to assist with technology assisted review of documents and the preparation of submissions and summaries are easily apparent. However, PGN 2/2023 highlights potential risks such as misleading information, potential breaches of confidentiality as well as bias. Any AI technology used in arbitration must be tailored to the legal industry and practitioners must understand the technology’s limitations. 

Translation: As stated above, the UAE operates a legal system through both English and Arabic. It is therefore imperative that all documents including evidence, submissions, and witness statements are officially translated into the language of the court by a certified translator. The inception of AI to automate translations would create an accessible legal environment while eradicating the delay and expense of translations. 

Speech Recognition: In practice, parties will jointly instruct a transcription service to transcribe the arbitral hearing. The service provider is usually an external resource which attracts additional expense. Recent amendments to the Arbitration Law oblige the arbitral tribunal to provide the parties with hearing transcripts. Where the arbitral tribunal is to provide this service, and while the parties will inevitably incur these costs, it is an attractive cost-saving option to adopt AI speech recognition into the hearing process. However, the technology used will require to meet the regulatory requirements of data protection and confidentiality in such a private forum. 

Decision Making: The arbitral institutions in the UAE contain similar provisions within the rules to ensure that an arbitrator is “independent and impartial at all times.9 AI could conceivably assist in future with the process of selecting an arbitrator in an impartial, data-driven way. 

A step further, and into the future perhaps, we could see the creation of an AI arbitrator. Currently, the arbitral rules within the UAE either include provisions that an arbitrator must be a “natural person10 or have qualifications or certain attributes.  Whether an award issued by an AI machine could ever be recognised and enforced at the local or international courts remains to be seen.  

Nevertheless, it is extremely likely that we will see the evolution of AI being deployed in arbitration proceedings in the UAE and Middle East region. 

 

This series will continue next week with the perspective from Australia.


1The United Arab Emirates (“UAE”) is a federation of seven emirates: Abu Dhabi, Dubai, Sharjah, Ajman, Fujairah, Ras Al Khaimah and Umm Al Quwain.

2 UAE-National-Strategy-for-Artificial-Intelligence-2031.pdf (ai.gov.ae)

3 Technology Innovation Institute | Advanced Technology Research Institute | International Center for Technology Research | TII

4 Technology Innovation Institute Announces Launch of NOOR, the World’s Largest Arabic NLP Model | Technology Innovation Institute (tii.ae)

5 Publications | Artificial Intelligence Office, UAE (ai.gov.ae)

6 DIFC enacts landmark regulation for autonomous and semi-autonomous systems: Clyde & Co (clydeco.com)

7 Data Protection Law DIFC Law No. 5 of 2020.

8 Intellectual Property Law DIFC Law No. 4 of 2019, as amended.

9 Article 15 of the ArbitrateAD Arbitration Rules 2024. A similar provision found at Article 14.1 of DIAC Arbitration Rules 2022. 

10 Article 10(1)(a) of Federal Law No. 6 of 2018 as amended. 
 

End

Stay up to date with Clyde & Co

Sign up to receive email updates straight to your inbox!

You might be interested in...