Over the past year, the fields of environmental law and legal ethics have experienced significant evolutions, spurred by ABA ethics opinions, rapid advancements in generative Artificial Intelligence (Gen AI), and heightened scrutiny of environmental, social and governance (ESG) reporting. This Year in Review article explores these developments and their implications for the practice of environmental law.
I. ABA Ethics Opinions
The ABA issued ethics opinions 509, 510, 511, 512, and 513. These opinions are general in nature, but they may be of interest to environmental, energy, and resource law practitioners.
A. Disqualification to Prevent the Misuse of “Confidential Government Information” (February 28, 2024)
Formal Opinion 509 discussed Model Rule 1.11. Legal practitioners are required to prevent the misuse of confidential government information by lawyers representing private clients. This obligation specifically refers to information obtained under governmental authority that the government is prohibited from disclosing. If a legal practitioner acquires such information while serving as a government officer or employee, they cannot represent a private client whose interests are adverse to the person about whom the confidential information was acquired, in a matter where the use of that information could materially disadvantage that person.
B. Avoiding the Imputation of a Conflict of Interest When a Law Firm is Adverse to One of its Lawyer’s Prospective Clients (March 20, 2024)
Formal Opinion 510 clarified Model Rule 1.18 and a legal practitioners’ duty to avoid conflicts of interest. Practitioners who are approached by a prospective client but are not retained are disqualified from representing another client with interests adverse to the prospective client in the same or a substantially related matter if the lawyer received from the prospective client information that “could be significantly harmful” to the prospective client. This opinion outlines reasonable measures including cautioning clients not to disclose excessive details and conducting necessary preliminary conflict checks.
C. Confidentiality Obligations of Lawyers Posting to Listservs (May 8, 2024)
Formal Opinion 511 cautions legal practitioners against sharing representation-related information on listservs. Practitioners should avoid posting even hypothetical scenarios if they have not obtained the client’s consent and there is any risk of revealing the client’s identity.
D. Generative Artificial Intelligence Tools (July 29, 2024)
Formal Opinion 512 discusses the use of generative AI tools by legal practitioners. The opinion emphasizes duties of competence, confidentiality and communication surrounding the use of this technology. Practitioners must understand the tools they are using and ensure their actions protect client data. It is also essential that practitioners maintain their own independent judgment when utilizing these tools and understand the limitations of Gen AI tools.
E. Duty to Inquire into and Assess the Facts and Circumstances of Each Representation (August 23, 2024)
Formal Opinion 513 discusses amendments to Model Rule 1.16, which requires legal practitioners to investigate and assess the facts and circumstances of each representation. The revisions to the rule reinforce the previously implicit requirement that legal practitioners be vigilant in their initial consultations to mitigate risks of facilitating criminal or fraudulent activities. The opinion emphasizes a risk-based approach, wherein legal practitioners manage the depth of their inquiry based on factors such as the potential client’s identity, the nature of the transaction and the relevant jurisdictions. The opinion also reminds practitioners that they are required to withdraw from representation if they possess actual knowledge that their services will assist in crime or fraud.
II. Navigating the Ethical Considerations of AI in Environmental Law
As discussed above, Formal Opinion 512 underscores the necessity for legal practitioners to maintain high ethical standards as they utilize Gen AI tools. Similarly, several state bar associations, including New York, Texas, Illinois, Florida, and California, continue to address the implications of AI in legal ethics through the issuance of formal opinions or guidance. An increasing number of judges are also beginning to request that attorneys submit certifications regarding their use of AI.
A. Duty of Competence
To provide competent representation, it is prudent for attorneys to maintain an appropriate level of technological competence in this dynamic landscape. ABA Model Rule 1.1 requires lawyers to “provide competent representation to a client.” Comment 8 to the Rule indicates that the requisite legal knowledge and skill includes keeping abreast of “the benefits and risks associated with relevant technology.” This raises questions about the extent of this duty in the context of AI. At a minimum, the duty of competency should obligate practitioners to possess a basic understanding of AI tools and their use in practice. In a comment on AI and legal practice, the American Bar Association’s Section of Science, Technology & Law stated that “attorneys that choose to use AI tools are under a competence obligation that would require them to understand the benefits and risks associated with that technology and, to the extent their obligations under Model Rule 1.4 require, to communicate that to the client.” Competency with AI extends beyond merely knowing how to use it; it encompasses the ability to explain how it works.
B. AI and the Attorney-Client Relationship
AI has the potential to revolutionize the attorney-client relationship by identifying client needs more efficiently and effectively. Model Rule 1.2 requires lawyers to maintain the integrity of the attorney-client relationship, stating, “a lawyer shall abide by a client’s decision concerning the objectives of representation and, as required by Rule 1.4, shall consult with the client as to the means by which they are to be pursued.” Some clients may be fearful of these emerging technologies and slow to trust that any risks outweigh the benefits AI may provide for their case. If an attorney has opted to integrate AI into their practice, must they abandon those resources if a client is uncomfortable? It is crucial that practitioners engage in open dialogue with their clients to address their concerns and explore alternative solutions. A conversation regarding a practitioner’s use of AI should ideally occur at the outset of any attorney-client relationship. For those already engaged, it becomes necessary to discuss the contemplated use of AI as soon as it becomes relevant to the client’s matters. Practitioners are advised to obtain a client’s informed consent, which entails providing a clear explanation of how AI will be used in their case. This includes the benefits and any potential risks. These conversations should allow a client to fully understand the implications of AI being used in their legal representation. In contrast, utilizing AI to ask general legal questions or assist with drafting involving the input of anonymous data should not require informed consent, as confidential client information is not being exposed to the tool.
C. Supervision of AI Tools
The increasing use of AI to assist attorneys in solving complex legal questions raises questions regarding obligations to supervise. Model Rule 5.3 states that “a lawyer having direct supervisory authority over the nonlawyer shall make reasonable efforts to ensure that the person's conduct is compatible with the professional obligations of the lawyer.”ABA Model Rules 5.1 and 5.3 obligate senior practitioners to oversee the work of junior lawyers and non-lawyer support team members. The 2012 change to ABA Model Rule 5.3 aimed to clarify the intent of the rule’s scope and seems to have expanded its reach to non-human assistance. The title of Model Rule 5.3 changed from “Responsibilities Regarding Nonlawyer Assistants” to “Responsibilities Regarding Nonlawyer Assistance,” implying that even technological assistance requires a lawyer’s supervision. But what constitutes supervision? It could entail a multitude of things, including verifying an AI’s work for accuracy, confirming legal theories, and ensuring that any suggestions generated by AI comply with ethical standards. However, adequate supervision could prove difficult if a lawyer is utilizing AI without a sufficient level of understanding of the technology. ABA Model Rule 5.3 also emphasizes that non-lawyers can never be used as a substitute for a lawyer’s services, so practitioners should think twice before abandoning their due diligence in the face of this technology.
D. Confidentiality, Data Security, and AI
Another aspect of the intersection between attorney-client relationships and AI involves the duty of confidentiality. Confidentiality and data security stand out as the foremost concerns for many when it comes to AI. Pursuant to ABA Model Rule 1.6, lawyers owe their clients a duty of confidentiality, stating that “[a] lawyer shall not reveal information relating to the representation of a client unless the client gives informed consent,” or if “the disclosure is impliedly authorized in order to carry out the representation.” Model Rule 1.6(c) further states that “[a] lawyer shall make reasonable efforts to prevent the inadvertent or unauthorized disclosure of, or unauthorized access to, information relating to the representation of a client.” Inputting client data into AI software exposes the information to the potential threat of being used by the software in its perpetual evolution.
E. Malpractice and AI
Delegating copious amounts of work to AI programs raises concerns about due process and malpractice. Overuse of these platforms has the potential to compromise both client standards and the integrity of legal standards. Conversely, failing to take advantage of AI could call into question the quality of an attorney’s service to their client. The use of AI to streamline tasks also presents an ethical question regarding attorney billing practices. Model Rule 1.5 states that “[a] lawyer shall not make an agreement for, charge, or collect an unreasonable fee or an unreasonable amount for expenses.” Clients may scrutinize their lawyers for overcharging them for manual tasks that could have been efficiently and cost-effectively tackled by AI. Attorneys may have an ethical obligation to bill less due to efficiency.
F. Environmental Sustainability of AI
Gen AI models require training that consumes significant amounts of energy. As environmental concerns around the use of this technology continue to mount, legal practitioners must determine whether the use of the technology aligns with their sustainability goals. Recent studies indicate that the use of AI to construct a 100-word email equates to the same amount of water as a standard water bottle. In 2019, researchers found that creating a generative AI model consumed the same amount of energy as a round-trip transcontinental flight for one person. It is estimated that preparing Chat GPT-3 to launch “consumed 1,287 megawatt hours of electricity and generated 552 tons of carbon dioxide equivalent, the equivalent of 123 gasoline-power passenger vehicles driven for one year.” Legal practitioners continue to weigh these sustainability concerns against the benefits of this innovative technology when advising clients. This concern is particularly significant for environmental law practitioners, as “neglecting to consider the negative effects alongside these potential benefits can lead to uncontrolled growth with lasting consequences on the environment.”