The adoption of Artificial Intelligence (AI) powered tools, like ChatGPT, has the potential to transform various industries and how tasks are performed. And many proponents argue that the legal industry is no exception. While the emergence of these technological advancements has the potential to deliver better client service and increase value, the risks of this new technology must also be considered. This Article highlights some of the ethical considerations for attorneys using AI. We will address some of the key concerns that attorneys must consider when employing AI in their legal practice.

BUT FIRST, WHAT IS AI 

AI has been defined as “cognitive computing” or “machine learning”. In a nutshell, AI tools are trained to perform tasks that typically require human intelligence. For example, ChatGPT, perhaps the most popular generative AI tool, is a large language model (LLM) that uses learning techniques to generate human-like responses to “prompts” or natural language inputs. In the legal context, AI can assist with an array of legal tasks including electronic discovery, litigation analysis, due diligence reviews, drafting of legal documents, legal research, and due diligence reviews to name a few.  

ETHICAL CONSIDERATIONS

Competence 

Under comment 8 to ABA Model Rule 1.1 and Texas Rule 1.01, competent representation requires an awareness of the “benefits and risks associated with relevant technology.”  

Application to the use of AI. In August 2019, the ABA adopted Resolution 112, which provided in part

Under Rule 1.1, lawyers also must have a basic understanding of how AI tools operate. While lawyers cannot be expected to know all the technical intricacies of AI systems, they are required to understand how AI technology produces results. As one legal commentator notes, “[i]f a lawyer uses a tool that suggests answers to legal questions, he must understand the capabilities and limitations of the tool, and the risks and benefits of those answers.

It’s essential to understand the limitations of these AI models. ChatGPT, for example, has a knowledge cutoff of September 2021. The model is unaware of anything that happened after that date, including recent caselaw and legislative and regulatory enactments. Additionally, there are numerous examples of  language models producing inaccurate answers. One example making the rounds on the internet is when SCOTUSblog asked about Ruth Bader Ginsburg’s most noteworthy opinions, ChatGPT incorrectly identified her as the author of a dissent in Obergefell v. Hodges.

Communication 

A lawyer’s duty of communication is also implicated. Under ABA Model Rule 1.4 and Texas Rule 1.03, the lawyer must reasonably consult with the client about the “means” by which the client’s objectives are to be accomplished. 

Application to the use of AI. ABA Resolution 112 provides: 

A lawyer’s duty of communication under Rule 1.4 includes discussing with his or her client the decision to use AI in providing legal services. A lawyer should obtain approval from the client before using AI, and this consent must be informed. The discussion should include the risks and limitations of the AI tool. In certain circumstances, a lawyer’s decision not to use AI also may need to be communicated to the client if using AI would benefit the client. 

Confidentiality

Under ABA Rule 1.6 and Texas Rule 1.05, lawyers are required to protect all client information from both intentional and inadvertent disclosure. Lawyers must take “reasonable efforts” to prevent inadvertent or unauthorized disclosure of client information.

Application to the use of AI. ABA Resolution 112 provides: 

To minimize the risks of using AI, a lawyer should discuss with third-party AI providers the confidentiality safeguards in place. A lawyer should inquire about “what type of information is going to be provided, how the information will be stored, what security measures are in place with respect to the storage of information, and who is going to have access to the information. AI should not be used in the representation unless the lawyer is confident that the client’s confidential information will be secure. 

When using a Generative AI model, lawyers are submitting information to a third-party server outside of their control over the internet, which runs the risk of violating a lawyer’s duty of confidentiality.  Unlike other research technologies (commercial legal databases or Google) that have privacy policies limiting the use of query information, Generative AI models may lack some of these privacy protections. For example, Open AI, the creator of ChatGPT, warns that prompts may be reviewed by their AI trainers to improve the product, that users are not able to delete specific prompts from their history, and users should not share sensitive information in their “conversations”. Accordingly, attorneys may use ChatGPT for general legal research and writing, but should not provide specific client details or confidential information.

Fees

An interesting question is whether an attorney must use AI if it would lower a client’s fees. ABA Model Rule 1.5 and Texas Rule 1.04 govern the reasonableness of attorney fees. Attorneys are prohibited from charging an unreasonable fee or unreasonable amount for expenses. Factors to be considered in determining reasonableness include “time and labor required, the novelty and difficulty of the questions involved, and the skill requisite to perform the legal service properly.”

Application to the use of AI. ABA Resolution 112 provides: 

In certain circumstances, a lawyer’s decision not to use AI also may need to be communicated to the client if using AI would benefit the client. Indeed, the lawyer’s failure to use AI could implicate ABA Model Rule 1.5, which requires lawyer’s fees to be reasonable. Failing to use AI technology that materially reduces the costs of providing legal services arguably could result in a lawyer charging an unreasonable fee to a client.

For example, if the use of AI could materially reduce time needed to perform legal research, prepare first drafts, or proof an agreement for defined terms and consistency (the bane of many first year’s existence), the failure to use AI may run the risk of charging an unreasonable fee.

Duty to Supervise 

Attorneys have the duty to supervise under ABA Model Rules 5.1 and 5.3, as well as Texas Rules 5.01 and 5.03

Application to the use of AI. ABA Resolution 112 provides: 

2012 ABA adopted an amendment to Model Rule 5.3 that changed the title of Rule 5.3 from “responsibilities regarding nonlawyer assistants” to “responsibilities regarding nonlawyer assistance.” The change clarified that the scope of Rule 5.3 encompasses nonlawyers whether human or not. Under Rules 5.1 and 5.3, lawyers are obligated to supervise the work of AI utilized in the provision of legal services, and understand the technology well enough to ensure compliance with the lawyer’s ethical duties. This includes making sure that the work product produced by AI is accurate and complete and does not create a risk of disclosing client confidential information.

There are some tasks that should not be handled by today’s AI technology, and a lawyer must know where to draw the line. At the same time, lawyers should avoid underutilizing AI, which could cause them to serve their clients less efficiently. Ultimately, it’s a balancing act. Given that many lawyers are focused on detail and control over their matter, it is easy to see why “the greater danger might very well be underutilization of, rather than overreliance upon, artificial intelligence.”

One consideration is how does an attorney possibly supervise an algorithm? As discussed under the Competence section above, AI has real limitations. Responses generated can be imperfect or problematic. Accordingly, review is absolutely essential and outputs must be verified before using them. 

CONCLUSION

The advent of AI-powered tools like ChatGPT has the extraordinary potential to reshape the legal industry, increasing efficiency and value, and producing cost savings. With all new technologies, we cannot only consider the benefits, but the risks as well. As attorneys integrate these tools, they must ensure they uphold their ethical obligations.