From Algorithms to Awards: Ciarb's New Guidelines on AI for Arbitration

Introduction

At a time where technology is rapidly transforming industries across the globe, the field of arbitration is no exception. The Chartered Institute of Arbitrators (Ciarb) has now taken a significant step forward by publishing a set of guidelines to address the use of artificial intelligence (AI) in arbitration proceedings (the 2025 Guidelines). These 2025 Guidelines aim to provide clarity and direction for practitioners and arbitrators around the world, ensuring that AI's potential to assist in dispute resolution procedures is harnessed effectively whilst simultaneously maintaining the integrity and fairness of arbitration proceedings.

Overview of the Guidelines

Ciarb’s 2025 Guidelines provide practical guidance and serve as a pioneering framework for the use of AI technologies within arbitration. They encompass several key areas:

Part I – Benefits and Risks of the Use of AI in Arbitration

The key benefits identified include enhancing the arbitral process in terms of efficiency and quality, as well as the ability to use AI related tools such as legal research, data analysis, text generation, translation, transcription and case analysis.

Alongside the benefits, the 2025 Guidelines also address the key risks associated with using AI such as potential complications surrounding the enforceability of arbitral awards, confidentiality, data security, and algorithmic bias.

Part II – General Recommendations About Use of AI in Arbitration

Ciarb recommends that parties and arbitrators make reasonable enquiries about prospective AI tools and the application of any AI-related rules in the relevant jurisdiction in which any arbitration is proceedings. Parties and arbitrators are also encouraged to engage in understanding the potential risks involved with the use of AI and weigh the balance of the benefits against such risks as opposed to adopting a blanket “one size fits all” approach.

Part III – Parties’ Use of AI In an Arbitration

Guidance on how the parties can effectively incorporate AI technologies into arbitration is provided, explaining how arbitrators can give directions and make procedural rulings on the use of AI, exploring the scope of an arbitrator’s authority to manage the use of AI by the parties.

Part IV – Use of AI by Arbitrators

How arbitrators themselves may utilise AI is also covered. The 2025 Guidelines make clear that whilst arbitrators should use AI to enhance the arbitral process and the quality of their decision making via the production of more accurate and efficient processing of the information submitted to the AI, they must take care not to relinquish decision making power to AI.

Appendix A and B

A template agreement on the use of AI in arbitration, and a template Procedural Order are also annexed for use.

Key Takeaways

There is no doubt that the 2025 Guidelines encourage practitioners and arbitrators alike to embrace AI and proactively adapt to the technological advancements that it offers. They also provide valuable clarity as to the use of AI in arbitration. As AI becomes more extensively used, they will act as a helpful reference point moving forward. By establishing starting point protocols and encouraging collaboration between the parties, their nominated experts and the appointed Tribunal, the 2025 Guidelines also strive to minimise disputes related to the use of AI within arbitration proceedings.

The 2025 Guidelines come at a time where there is a real need for legislation and guidance generally on the use of AI in law. The recent EU AI Act, which came into force on 1 August 2024 and will be fully applicable from 2 August 2026 to AI systems entering the market in EU states, is the first ever legal framework on AI worldwide and regulates the use of certain AI systems. During its two-year implementation period we will no doubt see other supporting legislation and guidance to supplement the Ciarb’s 2025 Guidelines. Watch this space.