Editor’s Note: The Fourth Seminar of the International Commercial Expert Committee of the Supreme People’s Court and Reappointment (New appointment) Ceremony of Expert Members was held successfully on September 25, 2024. Over 40 experts from more than 20 countries and regions focused on the theme of the “Collaborative Dialogue, Diverse Integration, Peaceful Development” during the seminar. Extensive and in-depth discussions were held within the framework of four specific issues. The texts of speeches delivered by the members of expert committee and distinguished guests during the discussion session on Topic Four Innovation in digital era: AI’s application in dispute resolution and its outlook would be posted on the CICC’s website.

Innovation In Digital Era：AI’s Application In Dispute Resolution And Its Outlook

Ismail Selim

Director of Cairo Regional Centre for International Commercial Arbitration

Board Member of the Africa Arbitration Association

I. INTRODUCTION

1. The topic ‘AI’s Application in Dispute Resolution and Its Outlook’ address the transformative impact of artificial intelligence (AI) on dispute resolution, particularly in arbitration and alternative dispute resolution (ADR).

2. It is important to note that there is no single definition of AI, and even existing definitions may evolve over time. According to the SVAMC Guidelines, AI refers to computer systems that perform tasks commonly associated with human cognition, such as understanding natural language, recognizing complex semantic patterns, and generating human-like outputs. The definition adopted is intended to be broad enough to encompass both current and foreseeable future types of AI, but not so broad as to include every type of computer-assisted automation tool. Rather, it focuses on modern technologies that tend to be more autonomous, complex, multifunctional, and probabilistic than traditional automation tools based on rule-based deterministic logic.

3. AI technologies, such as machine learning, generative AI, and data analytics, are increasingly being integrated into legal practices to enhance efficiency, reduce costs, and improve decision-making processes. As AI becomes more sophisticated, it holds the potential to revolutionize how disputes are managed, providing faster and more data-driven outcomes compared to traditional methods.

4. The aim of this note is to highlight both the opportunities and challenges posed by AI in dispute resolution. AI’s ability to automate routine tasks and analyze vast datasets can significantly streamline the arbitration process. However, the use of AI also introduces complexities, such as the need to ensure ethical standards, safeguard confidentiality, and address biases inherent in AI algorithms. These considerations are vital to maintaining fairness, transparency, and the overall integrity of the arbitration process.

II. GUIDELINES FOR AI USE IN ARBITRATION

5. The Silicon Valley Arbitration & Mediation Center (SVAMC) has developed a set of guidelines to standardize the use of AI in arbitration. These guidelines, known as the ‘SVAMC Guidelines on the Use of Artificial Intelligence in Arbitration,’ provide a structured approach to using AI responsibly and ethically in legal settings. They are intended for use by arbitrators, parties, legal representatives, and other participants in arbitration processes.

6. The SVAMC Guidelines are divided into three main parts: general principles applicable to all participants, specific provisions for parties and their representatives, and detailed rules for arbitrators. Key points include the duty of participants to understand the capabilities and limitations of AI tools, the obligation to protect confidential information, and the requirement that decision-making responsibilities must not be delegated to AI. Additionally, the guidelines address the importance of transparency, encouraging disclosure when AI significantly influences arbitration outcomes. Indeed, the key guidelines highlight:

· Understanding AI tools: Participants must familiarize themselves with the functionalities, limitations, and risks associated with AI applications. This understanding is crucial to ensuring that AI is used appropriately and that its outputs are critically evaluated.

· Safeguarding Confidentiality: Protecting sensitive information is a paramount concern. AI tools may retain data entered during use, potentially leading to breaches of privacy or confidentiality. The guidelines emphasize the need for careful vetting of AI tools to ensure they comply with data security and privacy standards, particularly when handling confidential arbitration materials.

· Non-Delegation of Decision-Making: Arbitrators are cautioned against delegating any decision-making responsibilities to AI. While AI can assist with data analysis and insights, final decisions must be made by humans to maintain the integrity of the arbitration process.

· Disclosure and Transparency: Disclosure of AI tools is not automatically required but should be considered case by case, balancing due process and privilege. Where necessary, arbitrators should provide details such as the AI tool used, its version, and the specific prompts or outputs involved. However, arbitrators must disclose any reliance on AI-generated information outside the record, allowing parties to comment and ensuring their right to be heard. This approach promotes transparency while ensuring AI outputs are critically evaluated.

III. CHALLENGES AND RISKS OF AI IN ARBITRATION

7. Several challenges and risks are associated with the integration of AI into arbitration and ADR. One of the most significant concerns is the potential for bias within AI models. AI systems often learn from historical data, which can embed existing biases into their outputs. This may negatively affect the right to a fair trial. Even if humans are naturally biased, they produce reasoned emotional, moral, logical, rational, and emphatic decisions. Therefore, users are urged to use AI tools with bias mitigation features and to critically review AI-generated outputs for fairness.

8. Another major challenge is the lack of transparency, often referred to as the ‘black box’ problem. Many AI systems, particularly those based on deep learning and neural networks, operate in ways that are not fully understandable to humans, making it difficult to ascertain how decisions are made. Furthermore, the phenomenon of ‘AI hallucination’ exists, where AI algorithms perceive patterns or objects that are nonexistent or imperceptible to human observers, creating, outputs that are inaccurate. This opacity complicates accountability and can undermine trust in AI-assisted arbitration outcomes. The concept of ‘explainable AI’ was discussed as a potential solution, providing insights into how AI systems reach their conclusions, thus enhancing transparency and oversight.

9. Data privacy and security are also critical issues. AI tools require access to large amounts of data to function effectively, which can include sensitive or confidential information. The SVAMC Guidelines recommend to all participants to ensure their use of AI tools is consistent with their obligations to safeguard confidential information (including privileged, private, secret, or otherwise protected data).

IV. AI’S IMPACT ON LEGAL PRACTICE AND ADR

10. AI’s impact on the legal profession extends beyond arbitration, fundamentally altering how legal profession conduct their work. Tools that automate document review, streamline legal research, and analyze large volumes of case law are becoming increasingly common, helping to reduce the time and cost of legal services. Predictive analytics, capable of forecasting case outcomes, are enhancing legal strategy development and client advisory services. However, the widespread adoption of AI in legal contexts also necessitates a shift in skills and knowledge. Legal professionals must appreciate that the outcomes produced by this technology will generate weighty legal and ethical issues, such as tort liability and criminal responsibility.

11. Looking ahead, AI will continue to evolve and become more deeply integrated into dispute resolution practices. The future lies in a decentralized approach to justice, based on wide global sharing of anonymized judicial data between all stakeholders. In the decentralized approach, judicial data will be controlled initially by their originators, the parties, and the judiciary, under state regulatory supervision. The data will be shared widely with developers of AI modules and other stakeholders who will compete among themselves in their services for the parties and courts, arbitration institutions and mediation centres.

12. An example of AI’s potential is ChatGPT, which can assist in various legal tasks. It can draft documents as procedural orders, and summaries of the facts, and handle logistical communications. It can facilitate party collaboration by translating legal documents, analyzing memorials, and organizing disputed and undisputed facts. ChatGPT can also predict the aftermath of an award by assessing historical cases, helping to predict the possible factual outcomes of their awards, and the chances of an award being annulled, denied enforcement or voluntarily complied with. Additionally, AI can assist in dialectic argumentation by identifying weaknesses in legal strategies and generating counter-arguments based on opposing party submissions. Finally, it could help represent parties who cannot afford legal counsel, making justice more accessible.

13. However, the advancement of AI in these areas must be guided by robust regulatory frameworks to address ethical dilemmas, ensure compliance, and protect the rights of all parties.

V. CONCLUSION

14. While AI presents significant opportunities for innovation in dispute resolution, its integration must be carefully managed. Establishing clear guidelines, ethical standards, and regulatory oversight will be crucial to ensure that AI is used in ways that enhance the efficiency, fairness, and transparency of arbitration and ADR processes.