The integration of artificial intelligence (“AI”) into all aspects of our lives has come on quickly. It started as a sci-fi experiment that may happen years from now, but has since shifted into an all-out transformation of technology at our disposal. Not only is AI infecting our personal lives, but we have underestimated how it has and will impact the legal profession. Now, AI systems are being implemented across all legal professions; whether you are an attorney, paralegal, reporter, administrator, or work with legal databases, AI has come and is likely to stay for the foreseeable future.
Law firms and the judicial system now rely on AI for day-to-day tasks, such as legal research, document review, administrative automation, and drafting documents. AI systems – such as LexisNexis’ “Protégé” – are able to sort through vast amounts of information, including case law, statutes, regulations, and secondary materials. Used properly, AI offers serious efficiencies in the legal workspace that cannot be ignored. Used negligently, it can lead to ethical violations, sanctions, and other risks to an attorney’s profession.
The Office of Legal Ethics Counsel and Advisory Committee of the Supreme Court of Missouri released an informal opinion (Missouri Informal Opinion 2024-11) on April 25, 2025, in regard to the use of AI in the legal profession. It requires the following:
- Competence (Rule 4-1.1): Attorneys must educate themselves on the types of generative AI available, understand the risks and benefits, and keep abreast of changes in practice, including technology.
- Confidentiality (Rule 4-1.6(a), (c)): Before using any generative AI platform, a lawyer must assess the platform’s terms of service, security measures, data usage policies, and data sources to ensure client information is protected. The lawyer must also make reasonable efforts to prevent inadvertent or unauthorized disclosure of client information.
- Verification: All AI-generated content must be reviewed and verified for accuracy before use; the opinion states that generative AI tools are not always accurate.
- Candor (Rule 4-3.3(a)): A lawyer must not submit AI-generated content containing false statements of material fact or law without verification.
- Court orders (Rule 4-3.4(c)): When courts have standing orders or local rules restricting or requiring disclosure of AI use, attorneys must comply.
- Professional independence (Rule 4-5.4): A lawyer cannot rely solely on AI-generated content; independent professional judgment must be maintained.
- Supervision (Rules 4-5.1, 4-5.3(c)): Supervisory attorneys must ensure subordinates and nonlawyer staff comply; partners who know of noncompliant conduct and fail to take remedial action bear ethical responsibility.
While the Office of Legal Ethics Counsel and Advisory Committee of the Supreme Court of Missouri has released this informal opinion governing the use of AI in the legal profession, two appellate cases show the potential sanctions that attorneys may receive if they misuse AI:
- Kruse v. Karlen, 692 S.W.3d (Mo. Ct. App. E.D. 2024): an opinion decided on February 13, 2024, the Eastern District Court of Appeals dismissed an appeal because 22 of 24 case citations in Appellant’s brief were fictitiously AI-generated citations. The Court awarded $10,000 in sanctions to Respondent because Appellant violated Missouri Supreme Court Rule 84.19, frivolous appeal.
- Stevens v. BJC Health Sys., No. ED112759 (Mo. Ct. App. E.D. 2025): an opinion decided on March 18, 2025, the Court addressed an appellate brief that contained six case citations that the court could not locate. The court warned that “AI in drafting legal documents may lead to sanctions if the party fails to perform a critical review of the end-product.”
Outside of Missouri, other national cases have dealt with similar issues. Such as Fletcher v. Experian Information Solutions, Inc., 168 F.4th 231 (5th Cir. 2026), where the United States Circuit Court for the 5th Circuit believed that normal sanction rules can govern AI-related misconduct. The Court found that while AI is a new technology that may be helpful, Plaintiff’s counsel filed a “reply brief on appeal containing numerous inaccurate citations, quotations, and statements of fact.” Id at 235. A show-cause order was issued to support the fabricated quotes and misrepresentations of law. In questioning of the Plaintiff’s attorney during the show-cause hearing, the attorney “misled, evaded, and violated her duties as an officer of this court” in failing to accept responsibility and be more forthcoming. Because of this, the court imposed a fine as a sanction for her conduct.
Lawyers and staff understandably want to take advantage of AI’s additional efficiencies. While AI is helpful, it is not perfect. If those in the legal profession want to utilize AI to make work more efficient, they can. However, that does not subvert the ethical and legal duties that attorneys owe to their clients and the court. Attorneys still owe ethical obligations that implicate the use of AI: proper training, confidentiality, and supervision, among others. AI’s use in the legal field should be used carefully, especially when used for legal research. As the cases above have shown, attorneys can, and will, be sanctioned for using hallucinated cases, as they should be. While using AI has made the profession more efficient in the realm of research and writing, attorneys must double-check the citations and ensure that the cases they are relying on truly exist.
AI is still new and evolving. AI will likely continue to offer additional efficiencies that can be immensely helpful to attorneys and staff. However, our ethical obligations still remain. AI cannot be used to replace the attorney or change the attorney’s role. We will likely continue to see additional cases regarding sanctionable AI use, all of which will likely stem from the same issue at the end of the day: attorneys owe a duty of candor and competence to the legal profession; AI does not change that.