Lawyers will remain responsible for their use of artificial intelligence to prepare court documents, but new rules are required to avoid AI interfering with the “integrity” of witness statements, the Civil Justice Council has recommended.
Its final report on how AI should be used for the preparation of court documents in civil proceedings – which followed an eight-month consultation – did not recommend any major new regulations.
Instead, it opted for a “proportionate, risk-based approach” that maintains professional responsibility while introducing “targeted safeguards” where needed, said Sir Geoffrey Vos, former master of the rolls, head of civil justice and chair of the CJC, who retired at the end of October.
AI has the potential to support the effective administration of civil justice, but its use must not “compromise the accuracy, reliability or integrity of material placed before the court”, Vos added.
The main suggested change is in how AI is used for written witness statements.
“Where a witness statement is that of a witness of fact, the witness should be required to make a specific declaration that the statement has not been prepared through the use of an AI tool that has generated or influenced its substantive content,” stated the CJC report, which was produced by a working group of lawyers, barristers and judges, including representatives from the Law Society and Bar Council.
Legal regulators should work together to provide guidance to the legal profession on the appropriate use of AI in the preparation of court documents, the CJC said. Guidance should be consistent with professional and ethical obligations, it added.
The report also recommended that the CPR should adopt a risk-based approach to AI that identifies the nature and consequences of using a particular AI tool. The CPR should also identify the risks to the “integrity of civil proceedings and the proper administration of justice”.
Legal professionals participating in the CJC’s consultation broadly agreed that no new rules were needed for their use of AI to prepare court documents.
However, concerns were raised about the “potential for AI tools to reshape, embellish or otherwise influence evidence in ways that may not be immediately apparent”, the report added.
The report comes as legal systems around the world attempt to maximise the benefits of using AI − quicker research, data analysis and preparation of court documents – and minimise its risks, most notably AI making up or ‘hallucinating’ legal information and arguments, including citations that are used in court.
Brett Dixon, Law Society president, welcomed the report’s broad recommendations, noting that the Law Society had contributed to the consultation.
But in an interview with Global Legal Post, Dixon said that legal professionals would need practical guidance on using AI for court documents.
How law firms implement the guidance on AI and put their systems in order is “where the real work will be done”, he said, adding that the Law Society will provide further guidance to its members on the matter.
He said: “[AI] is such a fast-moving, moveable feast that what people are really looking for is can somebody tell me what good looks like in relation to how I do [court-based legal] work, and that’s where membership organisations are at their best.”
Despite the report’s relatively modest proposals for rules on how AI is used in civil court cases, the practical consequences for lawyers could be “substantial”, said Reuben Vandercruyssen and Caroline Hearn of law firm Hogan Lovells Cadwalader in a note on the report.
AI is increasingly used in litigation workflows, including the preparation of witness evidence, they said. Revised guidance on AI use for courtroom documents would “add another layer of safeguards around expert evidence”, they added.
Countries including Singapore, New Zealand and the US have issued guidance on the use of AI in court proceedings. In some US jurisdictions, courts require attorneys to submit a certification or notice outlining any use of generative AI tools, especially for legal citations. They also need to confirm they have verified the information using reliable, traditional sources, the CJC noted in its consultation.
Last year, the United Nations Educational, Scientific and Cultural Organization (Unesco) published guidelines on the use of AI systems in courts and tribunals.
Email your news and story ideas to: [email protected]
