THE OKLAHOMA BAR JOURNAL 82 | SEPTEMBER 2024 Ethics & Professional Responsibility A Cautionary Tale By Richard Stevens LATELY, THERE HAS BEEN A great deal of publicity, information and misinformation about the use of generative artificial intelligence (GAI) by lawyers. While there is no one definition of artificial intelligence (AI), lawyers have used AI-based technologies in their practices for years. Lawyers use AI-assisted review in electronic discovery, contract analytics, basic research and other purposes. Recently released ABA Formal Opinion 512 discusses ethical considerations in the use of GAI, which creates various types of new content in response to questions posed by a user. The opinion attempts to identify ethical issues with the use of GAI tools, which it describes as “a moving target – indeed, a rapidly moving target.” The opinion is segmented into broad categories of ethics issues within single rules and groups of related rules. COMPETENCE: RULE 1.1 Rule 1.1 requires that a lawyer “provide competent representation to a client. Competent representation requires the legal knowledge, skill, thoroughness and preparation reasonably necessary for the representation.” Comment 6 makes clear that a lawyer must also “maintain the requisite knowledge and skill ... including the benefits and risks associated with relevant technology.” To do so, a lawyer “should either acquire a reasonable understanding of the benefits and risks of the GAI tools that they employ in their practices or draw on the expertise of others who can provide guidance about the relevant GAI tool’s capabilities and limitations.” Lawyers must also understand that GAI may produce inaccurate content. GAI tools create content taken from the internet or other proprietary sources. The GAI tool may produce inaccurate, unreliable, incomplete or biased results, depending on the quality of the source material the tool uses. The GAI tool may also lack the ability to understand the meaning of information that is used and may, as ABA 512 notes, also “combine otherwise accurate information in unexpected ways to yield false or inaccurate results.” ABA 512 also recognizes a fact that has been the subject of numerous news reports recently. “Some GAI tools are also prone to ‘hallucinations,’ providing ostensibly plausible responses that have no basis in fact or reality.” Uncritical reliance on such information may violate the duty to provide competent representation to a client under Rule 1.1. CONFIDENTIALITY: RULE 1.6 Under Rule 1.6, a lawyer is required to keep confidential all information relating to the representation of a client unless the client gives informed consent, disclosure is impliedly authorized to carry out the representation or disclosure is permitted by an exception. Self-learning GAI tools pose a risk that confidential information may be disclosed improperly. While a determination of the requirements of Rule 1.6 may be heavily fact-dependent, ABA 512 suggests that a lawyer consider the likelihood of disclosure or unauthorized access of confidential information along with the sensitivity of that information before inputting information relating to a client into a GAI tool. If informed consent is required, the lawyer must disclose specific information about the risk, client information to be disclosed, how that information may be used against the client’s interests and a clear explanation of the benefits of using the GAI tool. COMMUNICATION: RULE 1.4 ABA 512 provides guidance about the extent to which communication about the use of GAI may be necessary to comply with Rule 1.4. The opinion clarifies that the determination of whether and what to disclose will depend on the circumstances of each case. ABA Formal Opinion 512
RkJQdWJsaXNoZXIy OTk3MQ==