As a research lawyer with over 30 years experience, I often worry that advances to AI-assisted legal research tools will one day eliminate my job and that senior and junior lawyers will no longer seek my assistance with analyzing precedent case law to provide an answer on legal issues that arise in client matters. However, I am fortunate to work with a great group of colleagues and for a law firm that supports my legal research and knowledge management role.
Being engaged in my job is, of course, important, as it would be for any worker.
As stated by Chief Justice Dickson of the Supreme Court of Canada in Reference re Public Service Employee Relations Act (Alberta), 1987 CanLII 88 (SCC):
Work is one of the most fundamental aspects in a person's life, providing the individual with a means of financial support, and, as importantly, a contributory role in society. A person's employment is an essential component of his or her sense of identity, self-worth and emotional well-being.
In Lavoie v. Canada, 2002 SCC 23, Supreme Court of Canada Justice Bastarache stated: "...work is a fundamental aspect in a person's life, implicating his livelihood, self-worth and human dignity."
Although, if necessary, I could potentially pivot to perform other roles within the legal community and thereby retain a contributory role in society, cases continue to demonstrate that reliance on AI-assisted legal research remains dangerous to the legal profession and can undermine the justice system if lawyers and judges are not diligent in verifying the authenticity of cases presented to the court.
In Wadsworth v. Walmart Inc., (District of Wyoming, Case No. 2:23-CV-118-KHR), Justice Rankin was required to issue an order against three lawyers acting for the plaintiffs to show cause why they should not be sanctioned or otherwise disciplined for presenting fake cases to the court in connection with a motion the plaintiffs had brought before the court. In support of the motion, the lawyers had cited nine cases.
However, Justice Rankin explained: "The problem with these cases is that none existed...", with one exception.
Justice Rankin further explained that while some of the other cases cited led to actual cases, the actual cases were under different names.
The judge noted that one of the cases, purportedly determined by a Wyoming court in 2011, led to a Supreme Court of the Australian Capital Territory case.
Another case, also purportedly determined a Wyoming court, led to a federal court case.
Lastly, some of the cases cited by the lawyers led to no referred cases at all.
Justice Rankin found that the cases were not identifiable by a legitimate commercial source (Westlaw) and that the court was unable to locate apparent Wyoming cases in the court's local Electronic Court Filing System. However, lawyers for the defendants submitted that some of the mis-cited cases could be found on ChatGPT, raising the spectre that this AI platform was used by the plaintiffs' lawyers and that this AI tool was responsible for generating the fake cases.
Justice Rankin was also troubled by the language the plaintiffs' lawyers had used on the motion to explain "legal standards". For example, in connection with a federal rule of evidence, the lawyers stated that "Wyoming case law reinforced these principles" even though they cited (albeit fake) federal district court cases.
Accordingly, in the circumstances, Justice Rankin ordered that the plaintiffs' lawyers provide a true and accurate copy of all cases used in support of the motion to the court, except for the legitimate case. Alternatively, if they were unable to comply with this order, each lawyer was to separately show cause why they should not be sanctioned under the inherent power of the court to order sanctions for citing non-existent cases to the court.
If the cases cited could not be found, the court further ordered that each lawyer must provide a thorough explanation for how the motion and the fake cases were generated and explain their role in drafting and supervising the motion.
More and more people, including lawyers, are using open AI sources to help them understand legal issues or to support legal arguments. However, this form of "do-it-yourself" legal research is prone to providing false results and generating cases that do not exist.
This is not the first time that this has happened in the United States or in Canada. However in Ontario, the Rules of Civil Procedure were amended to require lawyers to certify the cases in their written legal arguments. Rule 4.06.1(2.1) states that a factum shall include a statement signed by the party's lawyer, or on the lawyer's behalf by someone the lawyer has specifically authorized, certifying that the person signing the statement is satisfied as to the authenticity of every authority cited in the factum.
Although a layperson might be excused for using ChatGPT to conduct legal research (which I would highly not recommend) and relying upon it, it is, in my view, inexcusable for a professionally trained lawyer, regardless of their level of experience, to cut corners and rely on the answers provided by the ChatGPT platform or any other AI platform to support a legal position.
At most, AI-assisted legal research is simply another tool in the legal research toolbox. AI-assisted legal research should be approached with caution, and should not be relied upon as a tool that will necessarily produce a legitimate end result to a legal issue.
The key takeaway from Justice Rankin's decision is that my job as a research lawyer is still valuable and may not be soon replaced by AI-assisted legal research.
At the time of the writing of this blog, it was unknown if the lawyers produced the copies of the cases requested by the court, or, if the cases did not exist, what sanctions were imposed against any of the lawyers. A PDF version is available to download here.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
