Since the widespread launch of cheap, accessible, generative AI platforms like Open AI’s ChatGPT, Bing Chat, Google Bard, Cohere Coral, etc. we’ve all read the news articles and blog posts written using AI to show off the possibilities these new tools present. In the legal industry, there’s buzz around the potential applications of AI when it comes to drafting, research, and streamlining simple tasks. However, one BC lawyer has demonstrated the risks as well.

The Case: Zhang v. Chen, 2024 BCSC 285

The Facts

The case itself dealt with an application for costs arising from an unsuccessful application brought by Mr. Chen. However, the case also dealt with the issue of whether special costs ought to be payable by Mr. Chen’s lawyer, Ms. Ke, for the inclusion of two non-existent cases in her notice of application.

Opposing counsel came to the conclusion the cases were fictitious after extensive research, which were subsequently discovered to have been invented by ChatGPT, and sought special costs against Ms. Ke for the delay and costs incurred to expose the AI “hallucination.”

Ultimately, the court did not impose costs against Ms. Ke, concluding that there was no intent to deceive, noting the sincerity of her apology to counsel and the court.

Takeaways

The Honourable Justice Masuhara, who presided over the matter, issued a reminder that lawyers are ultimately responsible for work product generated using “technology based solutions” and urged lawyers to review the content to ensure its accuracy. He also included a final comment, stating “As this case has unfortunately made clear, generative AI is still no substitute for the professional expertise that the justice system requires of lawyers.  Competence in the selection and use of any technology tools, including those powered by AI, is critical. The integrity of the justice system requires no less.”[1]

Experimenting with AI

As a fun experiment to test the reliability of AI, the following was written using ChatGPT, using the prompt “write a case summary of Zhang v. Chen, 2024 BCSC 285, focusing on the use of AI”:

Case Summary: Zhang v. Chen, 2024 BCSC 285 – The Role of AI in Legal Proceedings

In Zhang v. Chen, 2024 BCSC 285, the British Columbia Supreme Court grappled with the implications of relying on artificial intelligence (AI) in the context of legal proceedings. The case centered around a dispute between the plaintiff, Mr. Zhang, and the defendant, Mr. Chen, regarding a real estate transaction.

Mr. Zhang alleged that Mr. Chen failed to disclose material defects in a property sale, resulting in financial losses. During the trial, Mr. Zhang’s legal team utilized AI-powered software for document review and analysis. The AI software was tasked with sifting through a large volume of documents to uncover relevant evidence supporting Mr. Zhang’s claims.

However, despite the initial confidence in the AI’s capabilities, it became evident during the trial that the software failed to detect crucial evidence buried within the documents. This revelation cast doubt on the reliability and effectiveness of AI in legal practice, undermining the plaintiff’s case.

The defense argued that the plaintiff’s reliance on AI software deprived them of a fair trial, as they were unable to scrutinize the algorithm’s methodology or challenge its findings effectively. This raised concerns about transparency, accountability, and due process in the use of AI in legal proceedings.

Ultimately, Zhang v. Chen underscored the limitations of AI in the practice of law, particularly in cases requiring nuanced interpretation and human judgment. While AI can assist in streamlining processes and handling large volumes of data, it should not replace human oversight or critical thinking.

The case highlighted the need for legal practitioners to approach AI with caution and skepticism, recognizing its potential but also acknowledging its limitations. Furthermore, it emphasized the importance of establishing guidelines and standards for the responsible use of AI in the legal profession, ensuring transparency, fairness, and adherence to ethical principles.

In conclusion, Zhang v. Chen serves as a cautionary tale about the risks of overreliance on AI in legal proceedings. By maintaining a balance between the capabilities of AI and the expertise of human practitioners, the legal profession can harness the benefits of technology while upholding the integrity of the justice system.

A compelling narrative? Sure. Accurate? Not so much. Somehow, a family law matter became a real estate dispute. One of the parties became a lawyer. A trial emerged from nowhere, and the use of AI became about its ability to detect evidence.

AI in the Legal System

AI is a rapidly changing technology that law societies and courts alike are still grappling with. The Law Society of Ontario website has a handful of resources related to AI, but to date has not created specific rules and/or guidelines around their use.

In Canada, the Court of King’s Bench of Manitoba, the Supreme Court of Yukon, and recently, the Federal Court have issued practice directions and notices to the profession that require parties to advise the court if their materials were prepared using AI. The ability of AI to “hallucinate” and generate incorrect information specifically appears to be the driving concern at the heart of these directions.

[1] Zhang v. Chen, 2024 BCSC 285 at para 46.

de VRIES LITIGATION LLP is hiring! If you are a litigator with 5+ years of experience interested in practicing in estate, trust, and capacity law, check out the job posting here.