In a recent decision, Justice Griffin of the British Columbia Court of Appeal issued a frank warning to litigants who use artificial intelligence (AI) to assist them in preparing court materials. Justice Griffin noted that courts are increasingly encountering self-represented litigants who rely extensively on generative AI. Lawyers are also increasingly using AI at all stages of the litigation process, and should take heed of the Court of Appeal’s comments. This is particularly so given that lawyers will be held to a much higher professional standard when using generative AI.
In 1182573 B.C. Ltd. v. Gerber, 2026 BCCA 399 (1182573 B.C. Ltd.), the appellants sought to appeal an order of the BC Supreme Court ordering that they post security for costs and issuing directions for notices to admit. The respondents sought to quash the appeal on the basis that leave to appeal had not been sought or granted. The application to quash the appeal was dismissed. The notice of appeal was converted into an application for leave to appeal, which was then also dismissed. The appellants were not represented by counsel.
Justice Griffin’s reasons for judgment include some general guidance for litigants regarding the use of AI in preparing materials for court. Justice Griffin noted that the appellant’s pleadings, correspondence to the other side and his submissions all bore resemblance to AI-generated materials. Accordingly, Justice Griffin stated that she had “a few words of advice and warnings to give litigants about the use of AI-generated materials in court.”[1] They include the following:
- Risk of Hallucinations. The risk that AI will make up non-existent case authorities, known as hallucinations, is by now fairly well-known. According to Justice Griffin, if a litigant uses AI to assist them in finding cases, the tool they should use is CanLII, which offers a free AI search function which generates search terms based on plain language questions. No matter what tool a litigant uses, it is ultimately their responsibility to ensure that any cases put before the court are real and relevant. Any litigant that files submissions that include hallucinated cases is in violation of the Registrar’s Filing Directive (September 4, 2025), which states that litigants are “responsible for the authenticity and accuracy of all materials filed with the Court,” and risks a costs award being made against them.[2]
- Risk of Attaching Numerous and Irrelevant Legal Authorities. According to Justice Griffin, another risk of using AI is that the tool will generate long lists of case authorities that are not relevant to the issues before the court. Relying on long lists of irrelevant case law is “the opposite of persuasive advocacy” and makes submissions “incoherent.”[3]
- Risk of Bad Legal Advice. There is a risk that AI will give bad legal advice, which non-legally trained litigants will not recognize. According to Justice Griffin, “AI can be quite delusional about one’s legal rights and how to present legal argument.”[4] Justice Griffin also expressed concern regarding the tone taken in correspondence prepared using AI, noting that “AI has no moral code: no sense of ethics or honesty. Poor conduct by a party within litigation not only exacerbates the conflict and leads to more expense, but it can also undermine the person’s credibility. This can contribute to the court making negative findings against them, and a high costs award against them at the end of the case.”[5]
- Access to Justice Concerns. Finally, Justice Griffin noted that the use of AI to prepare court materials, without exercising independent and informed judgment, can result in multiple lawsuits and applications that have no merit. This increases the time and expense required to respond to these materials. It also increases the judicial resources required to review materials. According to Justice Griffin, “[j]udges are a limited human resource” and cannot provide an unlimited audience to litigants who have relied on AI to prepare court materials.[6]
Self-represented litigants are not alone in using AI in court. Lawyers increasingly have access to fiduciary-grade AI tools and are being encouraged to use them to assist with various tasks, including legal research, summarizing case law and even drafting pleadings. According to Justice Griffin, while the same risks outlined above apply to lawyers, lawyers who use AI can apply their professional training and judgment to guard against them.[7] Thus, while the majority of Justice Griffin’s comments in 1182573 B.C. Ltd. are focused on non-legally trained individuals, lawyers should take note of her warnings, particularly because they will be held to a much higher professional standard. The professional responsibility considerations for lawyers using generative AI have been enumerated by the Law Society of British Columbia in their Guidance on Professional Responsibility and Generative AI, and includes the duties of competence, confidentiality, honesty and candour, supervision of delegated work and information security, among others.
For more information on this topic, please reach out to the authors, Morgan Camley, KC, Samantha Chang and Kathryn Gullason.
[1] 1182573 B.C. Ltd. v Gerber, 2026 BCCA 399 at para. 80.
[2] 1182573 B.C. Ltd. v Gerber, 2026 BCCA 399 at paras. 81-84.
[3] Ibid at paras. 85-87.
[4] Ibid at para. 89.
[5] Ibid at para. 90.
[6] Ibid at para. 96.
[7] Ibid at para. 93.
