Artificial Intelligence (“AI”) is an extraordinary legal tool: it accelerates research, improves efficiency, and expands access to information. One could even argue that it has the potential to be the most impactful development on the practice of law since the development of internet-based legal research and email. Whether attorneys are prepared for it or not, AI has already changed how legal work gets done, and legal professionals can no longer ignore AI. But like every tool available for use in one’s legal practice, its value depends entirely on the skill, understanding, and judgment of the person using it. Thus, it is imperative that lawyers and law firms understand the power and limitations of AI, and that they commit themselves to using it competently and ethically, as the Rules of Professional Conduct require. This article addresses the recent developments in Maryland law regarding the ethical use of AI in the legal profession, and how attorneys should approach using AI tools.
AI tools can assist lawyers in several important ways, but they are far from perfect. One well-known and recurring issue associated with the use of generative AI tools is AI hallucinations, which can come in a few forms. For example, using generative AI programs for legal research can result in citations to cases that do not exist. Alternatively, AI programs may offer a citation to a real case but state a conclusion or holding that is wholly unsupported by the case. It was estimated in one study published in the Journal of Empirical Legal Studies in March 2025 that general-purpose large language model AI programs hallucinate on legal queries on average between 58% and 82% of the time!
As the legal profession incorporates AI into legal practice, some professionals have used AI technology without checking the substance of the results, requiring courts around the country to address irregularities that can appear in legal papers when an attorney does not diligently review AI-assisted work. These irregularities can include hallucinated cases and case citations, misquoted passages, citations to cases that do not support the legal proposition in the filed document, and more.
Mezu v. Mezu
The Appellate Court of Maryland recently addressed the impact such hallucinations in Mezu v. Mezu, 267 Md. App. 354 (2025). In fact, Mezu appears to be the first published Maryland appellate case to address AI hallucinations in papers filed with a court. In Mezu, a divorce case, the wife’s attorney submitted a brief with multiple fictitious cases, misquoted passages, and citations to cases that did not support the proposition for which they were cited. The Court noted the irregularities and issued a Show Cause Order to the attorney to show cause as to why he should not be sanctioned and/or referred to the Attorney Grievance Commission as a result. In response, the attorney explained that he had collaborated with his law clerk – a non-attorney – to draft the brief, and the law clerk utilized ChatGPT to search for relevant cases. The responsible attorney then reviewed and made edits to the brief, and while he claimed that he asked the law to ensure that the cases were accurately cited, the attorney did not personally verify any of the citations or that the cases cited even existed.
The Court noted that AI hallucinations that appear in court filed documents implicate several Maryland Rules. For example, Maryland Rule 1-311(b) states that an attorney’s signature on a document “constitutes a certification that the attorney has read the pleading or paper; that to the best of the attorney’s knowledge, information, and belief there is good ground to support it; and that it is not interposed for improper purpose or delay.” Signing a document, without performing due diligence as to veracity of the facts, allegations, and legal citations made in said document, violates this Rule. At the very least, according to the Court in Mezu, an attorney’s duties “require that attorneys read, and thereby confirm the existence and validity of, the legal authorities on which they rely.” Additionally, Maryland Rule 19-301.1 requires an attorney to provide competent representation to a client. The Appellate Court of Maryland held in Mezu, “[c]ompetent representation requires the legal knowledge, skill, thoroughness and preparation reasonably necessary for the representation.” Finally, Maryland Rule 19-305.3(b) provides that, “an attorney having direct supervisory authority over [a] non-attorney shall make reasonable efforts to ensure that the person’s conduct is compatible with the professional obligations of the attorney.” In other words, Attorneys cannot escape responsibility for problems caused by their employees, and the Appellate Court in Mezu held that the attorney’s supervision of his law clerk in that case did not meet the standard required under the Rules.
Having found that the attorney had violated several ethical rules, the Court then considered how it should respond by looking to how courts in other jurisdictions have handled similar issues. The Court noted that many courts have issued monetary sanctions against the offending attorney, referred attorneys for potential disciplinary proceedings, and some have ordered the attorney to serve a copy of the decision on the attorneys’ clients. In Mezu, the Court decided against issuing monetary sanctions because no such request had been made by the opposing party. The Court further decided not to allow the attorney to file a supplemental brief, as the case has been fully briefed and argued, and further delays would prejudice the opposing party. Feeling the weight of its ethical responsibility and considering the nature and severity of the conduct at issue, the Court ultimately decided to refer the attorney to the Attorney Grievance Commission.
Takeaways
There are several takeaways from Mezu that Maryland attorneys should consider. First, attorneys must always review papers filed with the courts and all documents and cases cited therein to ensure that the statements and citations in the filing are correct and exist. This verification is required whether AI tools are used or not. Second, if an attorney uses AI tools in connection with their practice, the attorney has an ethical obligation to diligently review all AI-generated work product to ensure its accuracy. There is no question that generative AI programs hallucinate cases, quotations, and holdings, among other things. Attorneys must remain vigilant and do the work necessary to ensure that all citations, propositions, and quotes are 100% accurate. Third, if an attorney fails to properly review AI-generated work product, he or she is likely to find themselves the subject of monetary and/or disciplinary sanctions. Other sanctions for submitting AI hallucinations may include the striking of filings before the Court, which could also lead to a claim of malpractice.
AI is a fantastic, innovative tool that will surely transform the legal profession forever. It is not a tool to fear, but rather, a tool to learn, harness, and control. Attorneys could and should learn how to integrate AI into their practice, but it is also an attorney’s obligation to do so ethically and responsibly.
What Does This Look Like?
First and foremost, attorneys should know what tools are available and how they can be used responsibly to assist their practice. Remember, some tools are better adapted for use in legal practice than others. In addition, there are significant confidentiality concerns when using AI that must be considered. Attorneys should become comfortable with AI as a tool, not a replacement for their experience, judgment, and review. Just as a responsible attorney has the duty to review an associate attorney’s work, attorneys must diligently review any AI output for consistency and accuracy. Lawyers should also take the reasonable steps necessary to educate themselves regarding this new technological companion through trainings, CLEs, and other available resources. If they do not, they run the risk of substantial monetary sanctions, or worse, a permanent stain on their disciplinary record.
Richard J. Berwanger, Jr. of Liff, Walsh & Simmons is a member of the firm’s Litigation practice and is the head of the firm’s Labor and Employment and Legal Ethics practices. He has years of experience advising both attorneys and law firms on legal ethics issues and defending attorneys before the Attorney Grievance Commission of Maryland and the D.C. Board on Professional Responsibility.
Liff, Walsh and Simmons’ Legal Ethics practice continually tracks developments in the field of legal ethics that may impact attorneys in Maryland or D.C. and their law firms. If you need any assistance developing AI or other ethics policies in your practice, addressing AI mishaps, or if you have any further questions, please contact Richard J. Berwanger, Jr., partner and head of the practice.
This alert provides general information and is not a full analysis of the matters discussed. It may not be relied on as legal advice. Richard J. Berwanger, Jr., a Liff, Walsh & Simmons partner licensed to practice law in Maryland and the District of Columbia.
