Arbitration In IndiaGuidelines for Artificial Intelligence in Arbitration

Introduction

The guidelines on the use of Artificial Intelligence (“AI”) in arbitration were introduced by the Silicon Valley Arbitration & Mediation Center (hereinafter referred to as “SVAMC”) on April 30, 2024 to provide a structured approach to integrating AI tools into arbitration proceedings responsibly.

Dispute Resolution is constantly being reshaped by AI and these guidelines are important for balancing innovation with ethical considerations, confidentiality, and decision-making integrity.

The rise of generative AI brings both advantages and challenges to arbitration at both national and international levels. While AI has the potential to enhance efficiency and accuracy by automating tasks like legal research, document analysis, translation, and drafting, it also raises important concerns regarding reliability, transparency, bias, confidentiality, and adherence to due process. The arbitration agreement adopted by the parties and/or the tribunal at any stage of the arbitral proceedings, should draft and incorporate such guidelines, either fully or partially.

This article aims to address the guidelines that should be a basic benchmark for jurisdictions all over the world.

Guidelines for all participants in Arbitration

• Understanding the uses, limitations, and risks of AI applications

Participants should make reasonable efforts to familiarize themselves with the functionality, limitations, and risks associated with AI tools used in the preparation or conduct of arbitration proceedings. Key considerations include:

1. “Black-box” issue: To enhance transparency, participants should prioritize using AI tools that provide insights into how a specific output was generated.
2. AI tools may not be ideal for tasks requiring domain-specific expertise or case-specific details unless they have been fine-tuned or supplemented with relevant data to enhance accuracy.
3. Errors: AI-generated content may sometimes be inaccurate and these errors can be minimized through techniques such as “prompt engineering” and “retrieval-augmented generation.”
4. Augmentation of Biases: Participants should exercise heightened caution when using AI tools for decision-making or assessments to mitigate the risk of bias.

• Safeguarding confidentiality

Participants need to make sure that using AI technologies complies with confidentiality requirements. AI systems shouldn’t receive confidential data without the appropriate authorisation. It’s critical to evaluate AI products’ data use and retention rules and choose ones that provide sufficient confidentiality protection.

• Disclosure of use of AI tools

Although disclosure of the use of AI technologies is not required in general, judgements on disclosure should be made case-by-case, taking into account pertinent facts, due process, and any applicable privileges. To aid in evaluation, information including the tool’s name, version, usage description, and input-output data may be shared when required.

Guidelines for Parties and Party Representatives

• Duty of competence or diligence in the use of AI

When employing AI technologies in arbitration, party representatives must abide with the relevant ethical guidelines and professional standards. They are accountable for any unfixed flaws in submissions and are in charge of checking AI-generated outputs for factual and legal accuracy.

• Respect for the integrity of the proceedings and the evidence

Parties, representatives, and experts should not use AI in ways that compromise the integrity of the arbitration or disrupt proceedings. This includes refraining from using AI to falsify evidence or mislead the tribunal or opposing parties.

Guidelines for Arbitrators

• Non-delegation of decision-making responsibilities

Arbitrators must not give AI tools any authority over their personal mandate, particularly when it comes to decision-making. The arbitrator’s independent evaluation of the facts, law, and evidence should serve as the foundation for the ultimate decision, even though AI can help with duties like data analysis and information management.

• Respect for due process

Arbitrators should not rely on AI-generated information outside the record without appropriate disclosure to the parties and providing them an opportunity to comment. Arbitrators should not presume that sources that can be independently verified exist or that the AI accurately represents them if the AI tool is unable to cite them.

AMLEGALS Remarks

The AI guidelines provide a valuable framework for many jurisdictions to follow for the responsible use of AI in arbitration, emphasizing the importance of understanding AI tools, maintaining confidentiality, ensuring accuracy, and upholding the integrity of arbitral proceedings. As AI technology evolves, these guidelines serve as a dynamic resource, adaptable to future advancements.

A key takeaway from these Guidelines is the emphasis on human oversight-AI can assist but cannot replace the judgment, reasoning, and responsibility of arbitrators and legal professionals. The Guidelines reaffirm that upholding professional and ethical commitments, protecting confidentiality, and using AI responsibly are crucial to preserving arbitration’s credibility.

Team AMLEGALS.

Please reach out to us at rohit.lalwani@amlegals.com in case of any query.