This browser is not actively supported anymore. For the best passle experience, we strongly recommend you upgrade your browser.

Technology Law

| 8 minute read

ABA Issues Comprehensive Formal Ethics Opinion on Lawyers’ Use of Generative AI

The American Bar Association (“ABA”) Standing Committee on Ethics and Professional Responsibility published its first Formal Opinion on lawyers’ use of generative artificial intelligence (GAI).  Formal Opinion 512, issued on July 29, 2024 (the “Opinion”), draws on previous ABA opinions on the use of new technologies in the practice of law, as well as more recent state bar association ethics opinions on GAI.  Since the Opinion was issued, other ethics committees, including the New York City Bar Association Committee on Professional Ethics have issued guidance on the issue.  See, e.g., New York City Bar Association Professional Ethics Committee Formal Op. 2024-5, available at: https://www.nycbar.org/reports/formal-opinion-2024-5-generative-ai-in-the-practice-of-law/. Among other things, the Opinion aggregates and expands on the ethics issues raised over the last several years as GAI gained visibility and use in legal practice.

The Opinion addresses multiple ethics rules implicated  by lawyers’ use of GAI under the ABA Model Rules of Professional Conduct (“MR”), including rules regarding competence (MR 1.1), confidentiality (MR 1.6, 1.9(c), and 1.18(b)), communication (MR 1.4), meritorious claims and candor toward the tribunal (MR 3.1, 3.3, and 8.4(c)), supervisory responsibilities (MR 5.1 and 5.3), and fees (MR 1.5).[1]

The Opinion’s analysis of a lawyer’s ethical obligations with respect to the use of GAI is summarized below. A few of the key takeaways are as follows: 

  • What is GAI? The Opinion describes the text-based GAI tools most relevant to legal practice as “prediction tools that generate a statistically probable output when prompted.”[2] This is a helpful working description of GAI as relevant to the legal industry, because it makes clear that the output of a GAI tool does not necessarily contain accurate or meaningful information, but a “statistically probable” output of fluent text based on a large amount of digital text it has culled and the particular prompt input into the tool. 
  • GAI is not a substitute for independent judgment. As the Opinion observes, at this point, “GAI tools lack the ability to understand the meaning of the text they generate or evaluate its context,”[3] and therefore are not a substitute for the independent professional judgment a lawyer must exercise in providing legal services or advice. The lawyer remains ultimately responsible for rendering competent legal services and ensuring compliance with the applicable ethics rules.[4]
  • Input matters. In addition to appropriate review, verification and judgment over a GAI tool’s output, the Opinion notes that lawyers should be aware of whether information put into a GAI tool could improperly end up in a later output, potentially compromising client confidentiality. This consideration requires a lawyer to obtain a reasonable level of competence before using a GAI tool, as well as a duty to evaluate whether an attorney must disclose the use of a GAI tool to a client and/or obtain informed consent as to use it.  Lawyers should also consider the implementation of appropriate policies or safeguards to manage and supervise the use of GAI tools. 
  • GAI is rapidly evolving. The Opinion notes that the duty to keep abreast of the technology’s benefits and risks “is not a static undertaking” in light of the “fast-paced evolution of GAI tools.”[5] While the Opinion strives to be comprehensive, the Opinion anticipates that the ABA as well as other state and local bar association ethics committees will have much more to say on the subject in future opinions as the technology changes,[6] signaling that lawyers should keep an eye out for new information, continue to update their knowledge and skill in using technology, and observe further ethical guidance as it emerges.

Competence, Meritorious Claims, and Candor to the Tribunal 

The Opinion notes that a lawyer’s duty of competence extends to their understanding of “the risks and benefits associated with relevant technology”, which they may achieve through self-study, associating with another competent lawyer or consulting with someone with sufficient expertise.[7] 

The Opinion identifies some of the benefits and risks presented by GAI – namely, the “potential to increase the efficiency and quality” of the lawyers’ work, compared to the risk of inaccurate, incomplete or discriminatory output. Since the lawyer remains ultimately responsible for providing competent legal services, the ABA observes that a lawyer’s uncritical reliance on GAI output without an “appropriate degree of independent verification or review” could violate the duty of competence.[8] What exactly constitutes an “appropriate degree” of independent verification or review depends on the GAI tool and the specific tasks for which a lawyer uses it (among other things).[9] For instance, using GAI to generate ideas “may require less independent verification or review.” Using a GAI tool to review and summarize long contracts would not require the lawyer to manually re-review each document, if the lawyer tested the tool’s accuracy by manually reviewing a smaller subset of documents summarized by the tool and found those summaries to be accurate. Overall, while lawyers may use GAI as a “springboard” for legal work, they “may not abdicate their responsibilities by relying solely on a GAI tool to perform tasks that call for the exercise of professional judgment”[10] and when considering employing a GAI tool, they “must consider the client’s interests and objectives.”[11]

Citing the same risks of unreliable or inaccurate output, the Opinion notes that lawyers should carefully review all GAI outputs in the context of litigation to ensure that the lawyer does not submit materials to a court that contain misstatements of law or fact, fail to include controlling legal authority, or make misleading arguments.[12]

Confidentiality and Communication 

The Opinion also cautions that using GAI could disclose confidential client information. The assessment will “depend on the client, the matter, the task, and the GAI tool used to perform it.”[13] The Opinion highlights self-learning GAI tools as raising the risk of improper disclosure because an input of confidential client information may be included in an output from a later prompt.  Consequently, the Opinion finds that a client’s informed consent is required before inputting their confidential information into a self-learning GAI tool, and that for the consent to be informed requires the lawyer’s explanation of the risk, not merely boilerplate provisions in an engagement letter.[14]

Relatedly, the Opinion describes when a lawyer must disclose their GAI practices, including upon client request or as required by the terms of the engagement agreement, as well as whenever the output of a GAI tool “will influence a significant decision in the representation.”[15] Again, the Opinion observes the necessity to disclose the attorney’s use of GAI is a fact-based analysis requiring lawyers to consider “the client’s needs and expectations, the scope of the representation, and the sensitivity of the information involved” as well as “the GAI tool’s importance to a particular task, the significance of that task to the overall representation, how the GAI tool will process the client’s information, and the extent to which knowledge of the lawyer’s use of the GAI tool would affect the client’s evaluation of or confidence in the lawyer’s work.”[16]

Supervisory Responsibilities 

The Opinion considers the duties of lawyers with managerial and supervisory responsibilities concerning the use of GAI by lawyers and non-lawyers under their supervision. It states that these lawyers must address GAI – managerial lawyers must establish clear policies regarding the firm’s permissible use of GAI, and supervisory lawyers must ensure lawyers and nonlawyers are trained in the ethical and practical use of relevant GAI tools and their risks, and ensure that they comply with their professional obligations when using those tools.  Finally, the Opinion draws on earlier opinions on cloud computing and outsourcing services to confirm that a lawyer’s obligation to ensure vendors or third party contractors will do the work capably and protect confidentiality apply also to third parties’ use of GAI tools.[17] In particular, lawyers should investigate the reliability, security measures, and policies of the GAI tool and ensure that the GAI tool is configured to protect the confidentiality and security of information, that the confidentiality obligation is enforceable, and that the lawyer will be notified in the event of a breach or service of process. 

Fees

Finally, the Opinion considers the effect of lawyers’ use of GAI tools on their duties to charge reasonable fees and expenses and to communicate the basis on which the lawyer will charge fees. Regardless of the fact that the use of GAI tools may allow lawyers to provide their legal services more efficiently, the Opinion confirms that lawyers who bill clients hourly must bill for their actual time and not more, further observing that it may be unreasonable under MR 1.5 for a lawyer to charge the same flat fee when using a GAI tool as when not using it if the GAI tool enables the lawyer to complete tasks much more quickly.[18] 

Considering whether the use of a GAI tool may be chargeable as an expense, the Opinion notes that to the extent a GAI tool works more as a piece of equipment, such as if it is embedded into word processing software, it should be considered overhead, and not charged to a client as an expense on a particular matter.[19] If, on the other hand a lawyer uses a third-party GAI service that charges on a per-use basis, it would be reasonable to bill for the out-of-pocket expense actually incurred for its use on the client matter. The ABA also applies this principle to the use of a proprietary GAI tool that a law firm may develop in-house, noting that a firm may agree in advance with the client about the specific rates to be charged for using the proprietary GAI tool, but absent such an agreement, may charge the client no more than the direct costs associated with providing the tool, with the appropriate disclosures.[20] Finally, unless a client specifically requests that the lawyer use a specific GAI tool with which the lawyer is unfamiliar, a lawyer may not charge a client for the time or cost of acquiring the requisite competence in how to use a specific GAI tool or service.[21]

 

The Opinion is a helpful guide for lawyers who are considering incorporating GAI into their practice.  As the Opinion remarks, lawyers should continue to observe the developments in technology and ethics, as more guidance has and will likely continue to follow in the wake of this rapidly evolving technology.


 

[1] While the Model Rules are not binding in any jurisdiction, most U.S. jurisdictions have adopted rules substantially similar to the Model Rules. As relevant here, the ABA’s analysis under the Model Rules may serve as helpful guidance as when considering a lawyer’s obligations under the corresponding rules of the state in which they practice. 

[2] Opinion at 1.

[3] Id. at 3.

[4] See id. at 4.

[5] Id.

[6] Id. at 2.

[7] See Opinion at 2-3 (citing MR 1.1 cmt. [8]).

[8] Id. at 3-4 (citing ABA Comm. on Ethics & Prof’l Responsibility, Formal Op. 08-451, at 1 (2008).

[9] Id. at 4.

[10] Id.

[11] Id. at 5.

[12] Opinion at 9-10.

[13] Id. at 6.

[14] Opinion at 6-7.

[15] Id. at 8-9.

[16] Id. at 9.

[17] See id. at 11.

[18] Opinion at 12.

[19] Id. at 13 (citing ABA Comm. on Ethics & Prof’l Responsibility, Formal Op. 93-379 at7.

[20] Id. at 14.

[21] Id.

The Opinion aggregates and expands on the ethics issues raised over the last several years as GAI gained visibility and use in legal practice

Tags

artificial intelligence, ai, generative ai, legal ethics, legal tech, technology law