International arbitration in 2025

The legal industry is no stranger to AI technologies and technology-assisted document review (aka predictive coding). However, the next wave of innovation promises even more specialized solutions tailored for legal professionals.

Some of the key opportunities include:

Legal research: AI can streamline legal research by analyzing vast datasets to uncover relevant insights. Platforms equipped with built-in databases, including arbitration-specific resources, enhance practitioners’ ability to prepare efficiently and comprehensively for cases.

Document review: technology-assisted review has been available for many years, with “active learning” algorithms continuously refining models as they process new documents, improving accuracy over time. In 2025 and beyond, next-generation AI is expected to elevate this capability, responding to complex queries across extensive document sets. For example, emerging review tools can identify a witness’s involvement in specific events or aspects of a case, digest interview memoranda, and compare witness statements for inconsistencies.

Drafting and summarizing documents: AI-powered tools can assist in the initial stages of drafting legal documents, including arbitral awards, pleadings, briefs, and contracts, by analyzing precedents and generating new content based on templates and user input. While practitioners must carefully review the outputs, the thoroughness of AI-generated drafts often surprises users. These tools also excel at summarizing lengthy documents, helping practitioners quickly identify key points and focus on critical aspects of a case.

Comparing arguments and predictive analysis: AI can compare arguments across different document sets, helping practitioners identify strengths and weaknesses in their cases. Predictive analysis tools can leverage historical data in an attempt to forecast a likely range of arbitration outcomes, enabling practitioners to make more informed decisions and develop effective strategies.

Case management and transcription services: AI-powered tools can manage entire cases, tracking deadlines, managing communications, and organizing documents, allowing practitioners to focus on more strategic tasks. Furthermore, AI can facilitate virtual hearings by providing real-time transcription, enhancing accessibility and creating clear records of proceedings.

While AI presents transformative opportunities for international arbitration, it also introduces critical challenges. Safeguarding data privacy and security is essential when handling confidential pleadings, privileged or commercially sensitive documents, and attorney work product. Ethical design and use of AI algorithms are crucial to prevent bias and discrimination. Despite its automation capabilities, AI requires careful oversight by practitioners to ensure nuanced legal analysis and sound decision-making. Practitioners must remain vigilant to avoid pitfalls such as a recent high-profile case in which AI “hallucinated” legal precedent by generating nonexistent case citations and quotes.

At present, there are no binding rules governing the use of AI in arbitration proceedings (subject to any jurisdiction-specific norms). However, 2024 saw the publication of several important guidelines and rules.

• The Silicon Valley Arbitration and Mediation Center released its Guidelines on the Use of AI in Arbitration, providing a principle-based framework for integrating AI tools into arbitration processes. They emphasize understanding the capabilities and limitations of AI, safeguarding confidentiality, and ensuring transparency and fairness in the use of AI.

• Similarly, the Stockholm Chamber of Commerce (SCC) issued a Guide to the Use of AI in Cases Administered under the SCC Rules, outlining best practices for incorporating AI into arbitration including: maintaining confidentiality, ensuring quality and integrity, and preventing the delegation of decision-making to AI tools.

• JAMS established comprehensive rules for handling disputes related to AI technologies.

Looking ahead, more arbitral institutions are likely to consider implementing rules governing the use of AI as its adoption expands. For instance, the International Chamber of Commerce (ICC) plans to launch a task force on AI in international dispute resolution. These efforts are essential to addressing concerns about security and confidentiality.

The regulatory landscape of AI is rapidly evolving, with varying approaches in different jurisdictions (for relevant updates, visit the Freshfields Technology quotient blog). For instance, the EU enacted the first comprehensive legal framework on AI, modeled on existing product safety laws. It includes a broad definition of AI and assigns various responsibilities to providers, deployers, and other stakeholders, with certain AI systems facing prohibition or strict regulation based on risk levels. US states such as California and Colorado have also been enacting their own AI regulations, the contours of which can vary significantly. Meanwhile, new AI laws have also been enacted in China, and proposed in key international arbitration hubs such as Brazil and Canada.

National legal organizations are also issuing guidance on AI use. Notably, the American Bar Association (ABA) released a formal opinion addressing the use of generative AI, emphasizing ethical obligations for lawyers and law firms. These include ensuring competent legal representation, safeguarding client information, and ensuring clear communication with clients and reasonable fee structures consistent with time spent using AI. Similar guidance is anticipated from other national bar associations.