OK Computer? CIArb Guidance on use of AI in Arbitration

The Chartered Institute of Arbitrators (CIArb) has recently released guidance on the use of AI in arbitration. The Guidance is a welcome tool that seeks to allow arbitration participants to harness the benefits of AI, whilst highlighting the risks of its use. 

The guidance is split into four parts, which consider:

The benefits and risks of the use of AI in Arbitration 

Here, the Guidance emphasises some of the clear advantages to using AI. Many of these benefits are oriented around the potential of AI to improve the efficiency and quality of the arbitral process, whether that be in legal research, data analysis or through the collection of evidence. 

The highlighted risks go to the heart of a number of reservations raised about AI generally, including:

• Issues associated with the input of confidential data into an AI tool, oriented around how such data is stored, used, protected and could emerge in response to a query raised by a third-party;
• The potential for bias in an AI tool based upon the configuration of an underlying algorithm and characteristics of a particular data set; and
• Concerns around due process; specifically possible inequality of arms (although AI could potentially help the otherwise under-resourced party) and discrepancies in parties’ ability to understand the case advanced against them.

General Recommendations about the use of AI in Arbitration

The focus of this section is on the importance of curiosity and enquiry, emphasising the importance for parties and arbitrators to make reasonable enquiry about an AI tool, understand the benefits and risks of using AI tools and consider AI-related laws, regulations and rules specific to the relevant jurisdiction. Perhaps most importantly, the Guidance emphasises a general rule that (unless agreed to the contrary) the responsibility and accountability of any participant in the arbitration process should not be diminished through the use of an AI tool.

Use of AI in an Arbitration

Here, the focus of the Guidance is on:

• The role of the arbitrator to (unless prohibited by applicable laws, regulations and procedural rules) regulate the use of AI by the parties through the use of directions and procedural rulings. Here, the arbitrator’s jurisdiction extends to the resolution of disagreements between the parties on the use of AI in the arbitral proceedings; 
• Autonomy of the parties to agree whether and how AI is used in the arbitration; and
• The importance of party disclosure as to their use of AI tools, “to the extent that its use may have an impact on the evidence, the outcome of the arbitration or otherwise involve a delegation of an express duty toward the arbitrators or any other party”.

Use of AI by Arbitrators

Here, the Guidance advocates that arbitrators may consider the use of AI tool to enhance the arbitral process and their own decision-making. However, crucially, the Guidance repeatedly advises that the decision-making powers of the arbitrator should not be relinquished to AI. Similarly, the Guidance advises against the delegation of tasks which may influence decision-making to AI tools.

Conclusion

This Guidance should be seen as a helpful reference point for those participating in the arbitration process as to how they grapple with the use of AI. Perhaps the most important points that we take from the Guidance are:

• Parties should welcome the use AI tools with eyes wide open;
• Consistent with the fact that arbitration is a creature of contract, the use of AI tools should be recorded in parties’ agreement and blessed by the arbitrator; 
• The use of AI tools should not diminish party responsibility and accountability; and 
• Decision making power should not be relinquished or delegated, in whole or in part, to AI tools.

For the latest on AI law and regulation, see our blog and newsletter.