Pretty much anyone who even knows a lawyer has a pretty good sense of how bad it was when the judge called out an attorney for submitting a legal document to the court that had made up cases and legal citations. But does this story warrant professional outcry? Demands for further rulemaking?
In my opinion, I do not believe whole cloth rules are needed to address the use of artificial intelligence (AI) in the practice of law. The American Bar Association has promulgated a set of Model Rules of Professional Conduct, and most states have adopted the model rules, in whole or in part. For example, ABA Model Rule 1.1 reads, “A lawyer shall provide competent representation to a client. Competent representation requires the legal knowledge, skill, thoroughness and preparation reasonably necessary for the representation.”
Similarly, the North Carolina State Bar’s Rule 1.1 states:
A lawyer shall not handle a legal matter that the lawyer knows or should know he or she is not competent to handle without associating with a lawyer who is competent to handle the matter. Competent representation requires the legal knowledge, skill, thoroughness, and preparation reasonably necessary for the representation.
North Carolina State Bar Rules of Prof’l Conduct 1.1 (2014).
Comment 8 to North Carolina’s Rule 1.1 further provides that an attorney’s competence includes “[t]o maintain the requisite knowledge and skill, a lawyer should keep abreast of changes in the law and its practice, including the benefits and risks associated with the technology relevant to the lawyer’s practice, . . . .” Id at Comment 8.
So, if it is known that AI systems have inherent flaws—outdated data packets as not all AI systems are connected to the internet, bias, and outright “hallucinations” of events that have never happened being returned with veracity being easy examples—attorneys have a heightened duty to understand the technology’s limitations and utilize it accordingly.
If you are of the mind that AI is completely different than any tech that has come before and, due to its complexity and ability to learn, is more akin to a person than other applications, the model rules still have us covered.
North Carolina has adopted the ABA’s Model Rule 5.3: Responsibilities Regarding Non-Lawyer Assistance as Rule 5.3: Responsibilities Regarding Nonlawyer Assistants. The North Carolina State Bar’s Comment 2 to Rule 5.3 includes:
A lawyer must give such assistants appropriate instruction and supervision concerning the ethical aspects of their employment, . . . and should be responsible for their work product. The measures employed in supervising nonlawyers should take account of the fact that they do not have legal training and are not subject to professional discipline.
North Carolina State Bar Rules of Prof’l Conduct 5.3, Comment 2 (2016).
Further, North Carolina State Bar’s Rule 5.5 details how attorneys who are not licensed in a jurisdiction should not practice law there nor should an attorney who is licensed aid a non-lawyer in the practice of law. While the rule does use the word “person,” if you consider an AI as a “person” at no time does an attorney run afoul of their professional obligations if they vet what the program produces based on the prompt. Just because AI spits out a response does not mean an attorney should shut their eyes and rely on it, much less file it with the court. I think most attorneys understand they should not do that if a paraprofessional or summer clerk just hands them something and tells them to sign it.
Under North Carolina law, anyone signing a document is charged with knowing its contents and having tested the veracity thereof.
The signature of an attorney or party constitutes a certificate by him that he has read the pleading, motion, or other paper; that to the best of his knowledge, information, and belief formed after reasonable inquiry it is well grounded in fact and is warranted by existing law or a good faith argument for the extension, modification, or reversal of existing law, . . . .
North Carolina General Statute § 1A-1, Rule 11(a) (1986).
In my opinion, at minimum, “reasonable inquiry” would include confirming that any facts or citations are true and accurate. And as the court stated in the now-infamous false citation sanctions order, ethics rules concerning lawyer conduct “impose a gatekeeping role on attorneys to ensure the accuracy of their filings."
Rather than avoiding artificial intelligence completely, attorneys should regard artificial intelligence as they would any other non-attorney staff. When an AI is gathering information from prospective clients to run conflicts checks and schedule appointments, the attorney needs to be mindful of confidentiality and conflicts of interest. If an AI is given a prompt to generate a legal argument in support of or opposition to an argument in 2,000 words or less, that attorney needs to review the output for clarity and cogency of the argument and run a citation check on any reference generated.
AI is a tool that, just like people, can transform the more administrative aspects of the practice of law when trained and supervised properly. But the human behind the keyboard needs to appreciate their ethical responsibilities to comprehend the tools they are using and supervise what is happening. Their professional reputation and ability to practice law depends on it.
