Generative AI Ethics in Missouri and Beyond

The most common ethical obligations implicated when working with Generative Artificial Intelligence (GAI) are (1) competence which includes understanding the risks/benefits of the GAI tool, (2) confidentiality which requires understanding the privacy policies of the GAI tool being used to protect client information and (3) candor to the tribunal which requires accuracy and in some jurisdictions, disclosure of the use of GAI in court filings.

ABA Formal Opinion 512

In July 2024, the ABA released Formal Opinion 512 to provide some guidance for lawyers regarding the use of GAI.

To comply with the ethical obligation of competence, lawyers do not need to become experts; but must have a reasonable understanding of the capabilities and limitations of the technology being used. When using a GAI tool, understanding the underlying data the tool is using to produce results is especially important because if that data is limited, outdated or biased in some way then the results produced may be unreliable.

The ABA stresses the importance of understanding the underlying data because GAI tools are prone to “hallucinations”. This is when the GAI tool provides plausible responses that have no basis in fact or reality.

To satisfy competence, the ABA recommends independent verification or review of GAI produced content along with consulting with others who are proficient in GAI technology to keep up with developments and understand how the technology is working.

Next, the ABA discusses the importance of maintaining confidentiality when using a GAI tool. At the time of this opinion, the ABA indicates there is a general lack of understanding about what happens to the information that is put into a GAI tool and how that information is protected. The ABA also highlights the possibility of inadvertent disclosures which can occur because a GAI tool uses the information put into it to improve its work in the future. The risks not only include disclosure to third parties but possible disclosure of confidential information within a single firm where the tool uses information from one client to generate information in a separate client’s case.

Due to these issues, the ABA requires lawyers to evaluate whether the use of a GAI tool will disclose confidential information or allows that information to be accessed by others outside the firm. The ABA recommends reading and understanding the products Terms of Use, privacy policy and any other related policies for the specific GAI tool and lawyers may even need to consult with IT professionals/cyber security experts.

Additionally, the ABA recommends clients give informed consent prior to putting their information into a GAI tool. Informed consent requires more than general boilerplate provisions in engagement letters and clients must understand why the tool is being used, the extent to the information being put in and the extent of the risk.

Lastly, the ABA discusses the use of GAI and candor to the tribunal where a common issue is inaccurate citations or using made up cases.

The ABA recommends that lawyers conduct an independent review of the results created by the GAI tool to verify it’s accuracy. There are also some courts requiring the disclosure of the use of a GAI tool as part of candor to the court.

Now let’s turn to Missouri, Kansas and Illinois to see what steps, if any, they have taken to address the use of GAI.

Missouri

Missouri does not have binding rules but was ahead of the ABA on releasing guidelines. On April 25, 2024, the Office of Legal Ethics Counsel & Advisory Committee of the Supreme Court of Missouri issued informal opinion 2024-11 which looks very similar to the ABA recommendations.

For competence, similar to the ABA, the Committee recommended that attorneys educate themselves on the types of GAI and understand the risk and benefits of using GAI.

The Office of Legal Ethics compared the use of GAI with work that is done with the assistance of a nonlawyer in which the lawyer has a professional responsibility to verify the accuracy of the work.

For confidentiality, similar to the ABA, the opinion states lawyers are required to make reasonable efforts to safeguard client information. The Committee recommends lawyers should consider the terms and conditions of a GAI platform to understand the security of the information prior to using GAI.

With respect to candor to the tribunal, see our article here about a recent case where the Eastern District of Missouri Court of Appeals stated citations to fake cases are a flagrant violation of the duty of candor.

Kansas

Kansas does not appear to have statewide rules yet, but Shawnee County has implemented District Court rule 3.125 Pleadings Using Generative AI. This rule states that if any portion of a pleading or filing was drafted using AI a litigant must (1) verify accuracy and (2) disclose to the court and opposing parties at time of filing that document contains AI content along with a certification that the filing was checked for accuracy. A violation of this rule may result in sanctions.

The requirement to disclose use of AI to the court is a way to comply with candor to the tribunal but neither Missouri or Illinois have implemented this requirement.

Illinois

In January 2025, the Illinois Supreme Court announced a general AI policy. The policy states that the rules of professional conduct apply fully to the use of AI, all users must review AI content prior to submitting to ensure accuracy, disclosure of AI use is not required, and AI applications must not compromise confidential information.

Conclusion