Master of the Rolls Predicts Humans Will Accept AI Judges for Small Disputes on Economic Grounds

Delivering the John Lehane memorial lecture to the Supreme Court of New South Wales on 18 August 2026, Sir Geoffrey Vos, the Master of the Rolls, predicted that humans will "come to accept, on economic grounds, machine-enabled resolution of small disputes" at first instance. Sir Geoffrey, who is retiring this autumn, argued the legal community "urgently" needed to debate where the line falls between "machine-made" and "machine-assisted" judicial decisions. Sir Geoffrey pointed to the growing use of LLMs (large language models) by judges and cautioned that, as confidence in tools such as Harvey or Legora grew, economic and time pressures could lead to judicial approval of machine recommendations without adequate scrutiny. He observed that responsible guidance issued by judiciaries globally already cautions that judges remain personally responsible for decisions whether AI-assisted or not, but acknowledged some judges may be tempted to use LLMs outside that guidance. As a concrete example of the direction of travel, Sir Geoffrey cited the American Arbitration Association's International Centre for Dispute Resolution, which announced last December an AI arbitrator for two-party, document-only construction disputes up to a certain value. That model uses a "human in the loop": the AI evaluates claims and prepares draft awards, but a human arbitrator reviews the reasoning and evaluates the decision to safeguard due process. Sir Geoffrey noted the Standing International Forum of Commercial Courts is working to draw up broadly agreed principles on what is acceptable in machine-made or machine-assisted decisions in commercial and arbitration cases. He said there was "probably no rule of law problem" with judges using AI for legal research, but flagged that the same tools are capable of answering the ultimate question in a dispute, creating a blurring of the line between assistance and decision-making.

Sign up to read →