Stepping Up to the Plate: Lawyer Duties in the Age of Generative AI

In our initial post in this series, we discussed the differentiation between AI overall and generative AI, and we’ll identify some of the current considerations impacting legal professionals in today’s landscape.

When it comes to best practices for AI ethics and defensibility in legal practice, rules and guidelines governing those best practices are key to informing lawyers and legal professionals how to behave ethically and defensibly when using AI.  If we listed all the currently proposed and issued rules and guidelines today, the list would be out of date almost immediately because state bars and courts are continually adopting new rule changes.

So, we’re going to address lawyer duties related to AI and generative AI differently here – by discussing six common duty themes associated with the various proposed and issued rules and guidelines by the American Bar Association (ABA), state bars and courts. In this post, we will discuss lawyer duties in the age of generative AI, with examples of rules that establish those duties.

Six Recurring Duty Themes for the Ethical and Defensible Use of Generative AI

Among the various proposed and issued rules and guidelines, six recurring duty themes organize nearly every AI authority.

Competence: In an extension of ABA Model Rule 1.1, Comment 8 regarding competence, lawyers have a practical obligation to understand the capabilities, limits, bias risks, confidentiality architecture, and hallucination risk of the particular tool being used. For example:

• The Alaska Bar Association’s Ethics Opinion 2025-1 says lawyers must understand “how the technology works”.
• The Arizona Bar Association’s Guidance for the Use of Generative Artificial Intelligence in the Practice of Law in Arizona says “Legal professionals must understand the functionality, limitations, and risks of generative AI before use”.
• The Illinois Supreme Court Policy on Artificial Intelligence says users must understand “both general AI capabilities and the specific tools”.
• Legal Ethics Opinion 24-01 from the Lawyer Disciplinary Board of West Virginia says AI should “supplement, not replace” legal reasoning.

Confidentiality: This is the sharpest operational risk. States repeatedly warn that public or consumer AI tools may store, reuse, share, or train on prompts and uploaded documents. For example:

• Ethics Opinion 24-1 of the Florida Bar requires lawyers to research “data retention, data sharing, and self-learning”.
• The Arizona Bar Association says a legal professional “must not enter any confidential client information into generative AI platforms unless adequate safeguards are in place.”
• Kentucky Bar Association Ethics Opinion KBA E-457 conditions nondisclosure in part on whether confidential information is exposed.
• The Illinois Supreme Court warns that AI “must not compromise sensitive information.”

Supervision: This applies both vertically and externally. Lawyers remain responsible for junior-lawyer work, nonlawyer staff work, and vendor/tool work when AI is involved. For example:

• The Washington State Bar Association’s Advisory Opinion 2025-05 explicitly lists supervision duties under RPCs 5.1 and 5.3
• The Office of Legal Ethics Counsel & Advisory Committee of the Supreme Court of Missouri’s Informal Opinion Number: 2024-11 identifies competence, confidentiality, candor, and supervision of nonlawyers.
• Arizona’s practical guidance treats AI deployment as something that must be governed by law-firm processes rather than left to ad hoc individual use.

Verification and Candor: The numerous case filings with AI hallucinations have made verification and candor the center of most court-facing AI rules. For example:

• Illinois’s Supreme Court policy says all users “must thoroughly review AI-generated content”
• The Pennsylvania Bar Association Committee on Legal Ethics and Professional Responsibility and Philadelphia Bar Association Professional Guidance Committee Joint Formal Opinion 2024-200 joint formal opinion says “lawyers have an obligation to verify that the
• citations are correct and that they accurately summarize the cases or other information cited”.
• The Supreme Court of Florida’s amended Rule 2.515(d)(2) now requires the signer to represent that “the legal authorities identified exist and are accurately cited”.
• There are also multiple federal standing orders that require party or counsel certifications to that effect.

Client Communication and Disclosure: These remain fact-sensitive rather than universally mandatory. In fact, several states reject a blanket rule that every use of AI must be disclosed to the client. For example:

• Kentucky says “there is no ethical duty to disclose the rote use of AI generated research for a client’s matter unless the work is being outsourced to a third party; the client is being charged for the cost of AI; and/or the disclosure of AI generated research is required by Court Rules.”
• Illinois’s policy says “disclosure of AI use should not be required in a pleading,” reflecting a similar functional approach in court filings.

Unauthorized Practice of Law: The advent of public AI chatbots has led to a significant increase in pro se filings. As a result, unauthorized practice of law (UPL) is increasingly framed as a “human in the loop” problem. For example:

• The Report and Recommendations of the New York State Bar Association Task Force on Artificial Intelligence states that “AI programs that do not involve a human-lawyer in the loop in providing legal advice arguably violate the rules and may be considered UPL”.
• The New Jersey Courts Notice on Use of Artificial Intelligence warns that “AI technologies might engage in activities that constitute the unauthorized practice of law, which is prohibited.”
• Many state sources emphasize that AI may assist but may not replace a lawyer’s legal judgment.

While there are variations from state to state, these six recurring duty themes provide a comprehensive set of expectations regarding lawyer duties regarding AI ethics.

ABA Formal Opinion 512, the American Bar Association’s primary ethics guidance for lawyers using generative AI in legal practice, mandates these same six duty themes as well – tying state ethics rules and guidance together.

Sources for AI Ethics Rules by State

It’s difficult to find a single source that has fully up-to-date AI ethics rules, so it’s important to check multiple sources to confirm that you are aware of the latest rules in your state. Here are two sources to check for AI ethics rules by state:

The Legal AI Compliance Register: Provides a state-by-state breakdown, with links to ethics rules and opinions, along with the last date the site was updated.

Justia AI and Attorney Ethics Rules: 50-State Survey: Provides a list of opinions or guidance from state bar associations regarding the use of generative AI.

Conclusion

As long as there continues to be regular misuse of generative AI and a general misunderstanding of how to use it by many in the legal profession, it will be important to emphasize rules and guidelines to inform lawyers and legal professionals how to behave ethically and defensibly when using AI. As most states have issued some level of guidance by now – some have issued multiple forms of guidance – it’s important to understand the rules and guidelines in your jurisdiction.

Next time, we’ll discuss real-world sanctions and cautionary cases involving fabricated citations and inaccurate filings generated by AI tools and how to avoid becoming a cautionary tale yourself.

For more regarding Cimplifi specialized expertise regarding AI & machine Learning, click here.