Formal Opinion 2024-5: Generative AI in the Practice of Law
Click “Download PDF” to access the full report
SUMMARY
The Professional Ethics Committee issued a formal opinion on the ethical obligations of New York lawyers and law firms using Generative Artificial Intelligence (“AI”) in practice. Because Generative AI tools are rapidly evolving, the guidance provided in the opinion is general. It follows and is consistent with the format used by the Practical Guidance for the Use of Generative Artificial Intelligence in the Practice of Law released by the California State Bar’s Standing Committee on Professional Responsibility and Conduct in November 2023. “Like the California State Bar and other bar associations that have addressed Generative AI, we believe that when addressing developing areas, lawyers need guardrails and not hard-and-fast restrictions or new rules that could stymie developments. By including advice specifically based on New York Rules and practice, this Opinion is intended to be helpful to the New York bar.”
REPORT
FORMAL OPINION 2024-5: ETHICAL OBLIGATIONS OF LAWYERS AND LAW FIRMS RELATING TO THE USE OF GENERATIVE ARTIFICIAL INTELLIGENCE
IN THE PRACTICE OF LAW
TOPIC: The use of generative artificial intelligence by New York lawyers, law firms, legal departments, government law offices and legal assistance organizations.
DIGEST: This opinion provides general guidance on the use of tools that use generative artificial intelligence.
RULES: 1.1, 1.2(d), 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 1.10, 1.11, 1.12, 3.1, 3.3, 5.1, 5.2, 5.3, 7.1, 7.3, 8.4
QUESTION: The availability of tools to assist lawyers in their practice that employ generative artificial intelligence has been dramatically expanding and continues to grow. What are the ethical issues that lawyers should consider when deciding whether to use these tools and, if the decision is made to do so, how to use them?
OPINION: When using generative artificial intelligence tools, a lawyer should take int account the duty of confidentiality, the obligation to avoid conflicts of interest, the duty of competence and diligence, the rules governing advertising and solicitation, the duty to comply with the law, the duty to supervise both lawyers and non-lawyers, the duty of subordinate attorneys, the duty to consult with clients, the duty of candor to tribunals, the prohibition on making non-meritorious claims and contentions, the limitations on what a lawyer may charge for fees and costs, and the prohibition on discrimination.
Introduction
Generative artificial intelligence (“Generative AI”), like any technology, must be used in a manner that comports with a lawyer’s ethical obligations. General-purpose technology platforms offer AI chatbots. Legal research platforms tout “legal generative AI” that can draft, analyze documents, and provide legal citations. Even data management vendors offer Generative AI-assisted review, analytic, and visualization capabilities. This summary of currently available tools will likely soon be outdated because of the rapid evolution of Generative AI. This guidance, therefore, is general. We expect that this advice will be updated and supplemented in years to come to cover issues not yet anticipated.
This Opinion provides guidance on the ethical obligations of lawyers and law firms relating to the use of Generative AI. It follows and is consistent with the format used by the Practical Guidance for the Use of Generative Artificial Intelligence in the Practice of Law released by the California State Bar’s Standing Committee on Professional Responsibility and Conduct in November 2023.[1] This Opinion is in the same format as the California State Bar’s guidance and contains multiple quotations from that guidance. Like the California State Bar and other bar associations that have addressed Generative AI,[2] we believe that when addressing developing areas, lawyers need guardrails and not hard-and-fast restrictions or new rules that could stymie developments. By including advice specifically based on New York Rules and practice, this Opinion is intended to be helpful to the New York Bar.
Click “Download PDF” to access the full report
Footnotes
[1] State Bar of Cal., Standing Comm. on Pro. Resp. & Conduct, Practical Guidance for the Use of Generative Artificial Intelligence in the Practice of Law (Nov. 16. 2023) (“California Guidance”), https://www.calbar.ca.gov/Portals/0/documents/ethics/Generative-AI-Practical-Guidance.pdf; see also Am. Bar Ass’n, Formal Op. 512 (2024); Fla. Bar Bd. Rev. Comm. on Pro. Ethics, Op. 24-1 (2024); D.C. Bar Ethics Op. 388 (April 2024); N.J. State Bar Ass’n, Task Force on Artificial Intelligence (AI) and the Law: Report, Requests, Recommendations, and Findings (2024), https://njsba.com/wp-content/uploads/2024/05/NJSBA-TASK-FORCE-ON-AI-AND-THE-LAW-REPORT-final.pdf; N.Y. State Bar Ass’n, Report & Recommendations of the New York State Bar Association Task Force on Artificial Intelligence (2024), https://www.nycbar.org/wp-content/uploads/2024/06/20221290_AI_NYS_Judiciary.pdf. (All websites last accessed on Aug. 5, 2024).
[2] In general, this Opinion is consistent with the ABA, California Bar, Florida Bar, District of Columbia Bar, and New Jersey Bar opinions cited in Footnote 1. However, the New York State Bar suggests adoption of certain rules to address Generative AI, which we believe is premature because of the rapid pace of technological development and change. See, e.g., N.Y. State Bar Ass’n, supra, at 53–56.
[3] Generative AI systems that share inputted information with third parties are sometimes called “open” systems.
[4] California Guidance at 2.
[5] Lawyers may wish to obtain advance client consent to use Generative AI that will involve sharing of client information, but, because such consent must be knowing, the client must understand the potential consequences of such information-sharing for the consent to be effective. See N.Y. State Op. 1020 ¶ 10 (a lawyer “may post and share documents using a ‘cloud’ data storage tool” that does not provide “reasonable protection to confidential client information” only where “the lawyer obtains informed consent from the client after advising the client of the relevant risks”).
[6] Id.
[7] See Am. Bar Ass’n, Formal Op. 512 at 6-7 (2024).
[8] Consistent with Rule 1.0(h), in this Opinion “law firm” includes a private firm as well as qualified legal assistance organizations, government law offices and corporations, and other entities’ legal departments.
[9] California Guidance at 2.
[10] Id.
[11] See N.Y. State Bar Ass’n, supra, at 58.
[12] California Guidance at 2. There have been claims that certain Generative AI tools violate intellectual property rights of third parties. A lawyer planning to use a Generative AI tool should keep abreast of whether there are any such risks associated with the tool the lawyer plans to use.
[13] Id.
[14] Id. at 3.
[15] Id.
[16] Id.
[17] Id.
[18] Likewise, where a client provides citations to a lawyer, a lawyer must review the decisions to make sure that they are genuine and properly cited. See United States v. Cohen, No. 18-CR-602, 2024 WL 1193604 (S.D.N.Y. Mar. 20, 2024) (criticizing an attorney-defendant and his counsel for citing “three cases that do not exist” where client provided citations hallucinated by Google Bard and counsel failed to check them).
[19] See Fla. Bar Bd. Rev. Comm. on Pro. Ethics, supra (section on Oversight of Generative AI).
[20] California Guidance at 4.
[21] Note that some Generative AI is routinely used. For example, Microsoft Word employs Generative AI in its auto-complete and grammar check functions. Westlaw, Lexis, and search engines also employ Generative AI. We do not mean to suggest that an attorney needs to disclose such uses of Generative AI. For a discussion of the importance of evaluating Generative AI tools based on intended users, see N.J. State Bar Ass’n, Task Force on Artificial Intelligence (AI) and the Law: Report, Requests, Recommendations, and Findings 15–19 (2024) (discussing “AI Tools Intended for the Public” and “Tools Tailored for Legal Professionals”), https://njsba.com/wp-content/uploads/2024/05/NJSBA-TASK-FORCE-ON-AI-AND-THE-LAW-REPORT-final.pdf.
[22] A Stanford University study found that Generative AI chatbots from OpenAI, Inc., Google LLC, and Meta Platforms Inc. hallucinate “at least 75% of the time when answering questions about a court’s core ruling.” Isabel Gottlieb & Isaiah Poritz, Popular AI Chatbots Found to Give Error-Ridden Legal Answers, Bloomberg L. (Jan. 12, 2024), https://news.bloomberglaw.com/business-and-practice/legal-errors-by-top-ai-models-alarmingly-prevalent-study-says. Courts are already grappling with parties’ citation to hallucinated precedents. See generally Mata v. Avianca, Inc., No. 22-CV-1461, 2023 WL 4114964 (S.D.N.Y. June 22, 2023) (sanctioning attorneys for “submit[ing] non-existent judicial opinions with fake quotes and citations created by the artificial intelligence tool ChatGPT”); Cohen, 2024 WL 1193604; see also D.C. Bar, Ethics Op. 388 (2024) (discussing the dangers of hallucinations).
[23] California Guidance at 4.
[24] See N.Y. City Op. 2018-4 (discussing a lawyer’s duty to inquire when asked to assist in a transaction that the lawyer suspects may involve a crime or fraud); see also ABA Op. 491 (2020); Colo. Bar Ass’n Ethics Comm., Formal Op. 142 (2021). These same standards apply when a lawyer suspects that a client may have given the lawyer fabricated evidence.
[25] See Rule 3.3.
[26] California Guidance at 4.
[27] Id.
[28] Id.
[29] See ABA Op. 93-379 (1993).
[30] California Guidance at 4.