NCSB Journal Summer 2026

THE NORTH CAROLINA STATE BAR

SUMMER 2026 JOURNAL

IN THIS ISSUE Guarding Against AI Errors page8 Why Your Firm Needs an AI Policy page 14 Finding the Next Starfish page19

FREE for North Carolina State Bar members, Smokeball Bill helps law firms manage trust accounting compliantly, bill easily, and get paid faster. Claim Your FREE Smokeball Bill Trust Accounting & Billing Software

or visit: smokeball.com/northcarolinabill Scan to get FREE Smokeball Bill

Smokeball is the trusted provider of legal practice management software to small and mid-size firms. Use of Smokeball does not ensure compliance with trust accounting rules. The North Carolina State Bar does not endorse, recommend, guarantee, or make any warranties regarding Smokeball products or services and disclaims any liability arising from their use. Attorneys should independently assess the risks, benefits, and limitations of any product or service.

THE NORTH CAROLINA STATEBAR JOURNAL Summer 2026 Volume 31, Number 2 Editor Jennifer R. Duncan © Copyright 2026 by the North Carolina State Bar. All rights reserved. Periodicals postage paid at Raleigh, NC, and additional offices. POSTMASTER: Send address changes to the North Carolina State Bar, PO Box 25908, Raleigh, NC 27611. The North Carolina State Bar Journal invites the sub mission of unsolicited, original articles, essays, and book reviews. Submissions may be made by mail or email (jduncan@ncbar.gov) to the editor. Publishing and editorial decisions are based on the Communications Committee’s and the editor’s judgment of the quality of the writing, the timeliness of the article, and the potential interest to the readers of the Journal .The Journal reserves the right to edit all manuscripts. The North Carolina State Bar Journal (ISSN 10928626) is published four times per year in March, June, Septem ber, and December under the direction and supervision of the council of the North Car olina State Bar, PO Box 25908, Raleigh, NC 27611. Member rate of $6.00 per year is in cluded in dues. Nonmember rates $10.73 per year. Single copies $5.36. Advertising rates available upon request. Direct inquiries to Editor, the North Carolina State Bar, PO Box 25908, Raleigh, North Carolina 27611, tel. (919) 828-4620. ncbar.gov Follow us at: X: @NCStateBar Facebook: facebook.com/NCStateBar

F E A T U R E S 8 Guarding Against AI Errors: Ethical Risks for NC Attorneys By Fred W. DeVore III and Rob Wilder 14 Beyond the Ban: Why Your Law Firm Needs a Realistic AI Policy in 2026 By Catherine Reach 16 Meet Union County’s Mother Daughter Public Defender Duo By Amanda Bunch 19 Finding the Next Starfish By Benjamin R. David 22 Shaping the Legal Landscape: North Carolina’s Role in the American Inns of Court Movement By Jason Hicks

Publication of an article in the Journal is not an endorsement by the North Carolina State Bar of the views expressed therein.

3

THE NORTH CAROLINA STATE BAR JOURNAL

DEPARTMENTS 5 President’s Message 6 State Bar Outlook 24

34 36 45

47 48 49 54

IOLTA Update Rule Amendments

In Memoriam

Distinguished Service Award July Bar Exam Applicants

Ethics Committee Update

Lawyer Assistance Program

Random Audit

28 30 31

BAR UPDATES 27

Legal Specialization

Upcoming Appointments

The Disciplinary Department

47

Client Security Fund

Pathways to Well-Being

Officers Katherine A. Frye, Raleigh President 2025-2026 Kevin G. Williams, Winston-Salem

Executive Director Peter G. Bolac

20: Joshua D. Malcolm, Pembroke 21: F. Davis Poisson III, Wadesboro 22: Judge Jason E. Ramey, Wentworth 23: Judge Tom Langan, Pilot Mountain 24: Judge Patrice A. Hinnant, Greensboro Manisha P. Patel, Greensboro 24H: Kathleen E. Nix, High Point Fred W. DeVore III, Charlotte M. Heath Gilbert Jr., Charlotte George V. Laughrun II, Charlotte Craig T. Lynch, Charlotte Gena Graham Morris, Charlotte Judge Nancy Black Norelli, Charlotte Timika Shafeek-Horton, Charlotte 27: T. Gregory Jones, Salisbury 28: Bobby Khan, Monroe 29: Matthew Rothbeind, Carthage 30: Judge Nathan Hunt Gwyn III, Monroe 31: Judge George M. Cleland IV, Winston-Salem 32: Daryl G. Davidson Sr., Taylorsville 33: Shawn L. Fraley, Lexington 34: John G. Vannoy Jr., North Wilkesboro 35: L. Jasmine McKinney, Boone 36: Timothy J. Rohr, Lenoir 37: Judge Brad V. Long, Asheboro 38: Michael R. Neece, Gastonia 39: Ret. Judge Jim Morgan, Shelby 40: Gill Beck, Asheville 41: Merrimon B. Oxley, Forest City 42: Michael A. Lovejoy, Hendersonville 43: Zeyland G. McKinney, Murphy Public Members Patricia Head, Littleton Samantha McKenzie Holmes, Morrisville Damon Seils, Carrboro 25: Amanda Reed, Concord 26: Stacy C. Cordes, Charlotte

Associate Director and Chief Ethics Counsel Brian P.D. Oten Counsel Carmen H. Bannon Editor Jennifer R. Duncan Communications Committee Damon Seils (Public Member), Chair Judge Tom Langan, Vice-Chair Thomas D. Anglim Julie Beavers (Advisory Member) Judge Takiya Lewis Blalock Ashley Campbell (Advisory Member) Andrea Capua (Advisory Member) Judge George M. Cleland IV Amy Davis (Advisory Member) Margaret Dickson (Advisory Member) Dionne Loy Fortner M. Heath Gilbert Jr. Patricia Head (Public Member) Samantha McKenzie Holmes (Public Member) T. Gregory Jones Gena Graham Morris Kathleen E. Nix Judge Ashleigh S. Parker F. Davis Poisson III Bill Powers (Advisory Member) Judge Jason E. Ramey Brandon A. Robinson David L. Sherlin G. Gray Wilson (Advisory Member)

President-Elect 2025-2026 David N. Allen, Charlotte Vice-President 2025-2026 Matthew W. Smith, Eden Past-President 2025-2026

Peter G. Bolac, Cary Secretary 2025-2026 Councilors By Judicial District 1:

John D. Leidy, Elizabeth City Tom D. Anglim, Washington Judge Jeffrey B. Foster, Greenville Scott C. Hart, New Bern Kevin Joseph Kiernan, Clinton Will Oden III, Wilmington Judge Takiya Lewis Blalock, Ahoskie

2: 3: 4: 5: 6: 7: 8: 9:

Michael B. Peters, Tarboro

J. Mark Herring, Kinston 10: Judge Julie L. Bell, Raleigh Anna Davis, Raleigh Judge Keith Gregory, Raleigh Jessica Heffner, Raleigh Kimberly A. Moore, Raleigh Judge Ashleigh Parker, Raleigh David Sherlin, Raleigh 11: J. Thomas Burnette, Oxford 12: Heather Williams, Dunn 13: Dionne Loy Fortner, Smithfield 14: H. Lee Boughman Jr., Fayetteville 15: Kyle Melvin, Elizabethtown 16: Meredith Nicholson, Durham Brandon Robinson, Durham 17: Kristen DelForge, Burlington 18: Fran Muse, Chapel Hill

SUMMER 2026

4

THE PRESIDENT’S MESSAGE

The Privilege We Inherited—and the Responsibility We Must Uphold B Y K A T H E R I N E F R Y E

In my speech the night I was sworn into this role, I talked about how I loved the law and loved being a lawyer. Being a lawyer is a privilege. As I have spent the last quarter of my presidency dealing with unexpected

ourselves to standards that protect the pub lic, the courts, and the legal system. Our Rules of Professional Conduct say it plainly: self-regulation “helps maintain the legal pro fession’s independence from government

A trustworthy disciplinary system and a commitment to using our professional obli gations for the public good are not just nice features of self-regulation, they are proof that we are earning the public’s trust. Again, What Do We Need to Do Today? Self-regulation has never been uncondi tional. When our self-regulation process comes under scrutiny, the right response is not to get defensive. It is to be transparent, engage seriously, and show that we are put ting the public interest first. For me, the last quarter has been about explaining and some times defending these programs or defending our right to self-regulate. A disciplinary process that becomes less transparent does not protect self-regulation. When the interest generated by client trust funds cannot reach the people it is meant to serve, we must ask ourselves what we are going to do about it. Communities that were already legal deserts are losing what little access they had. Real people are showing up to courthouses without the help they used to be able to get. Hopefully, by the time you read this arti cle, some of these issues will have been resolved. But keep in mind that whether they have been or haven’t been, everyone must remember how we earned the privilege to self-regulate and must remember that it is not guaranteed tomorrow. I implore anyone reading this article to ask how you can be proactive today to protect this privilege: Can you write a state legislator to explain the importance of IOLTA or explain the disci plinary process? Can you volunteer with a group that helps improve the court system? Can you accept a pro bono case? All of us are needed to protect this privilege. n

issues surrounding the disci pline of lawyers and the con tinued freeze of IOLTA funds, I have also realized how easy it is to take for granted that we regulate our selves. That is a privilege— and right now, it is one that deserves our full attention.

domination,” and “an inde pendent legal profession is an important force in preserving government under law.” What Do We Need to Do Today? The privilege of self-regu lation is something we have to keep earning. That means holding ourselves and each other to high standards of competence, ethics, and pro fessionalism. It means run

A Privilege Earned

I will be the first to admit that history was never my strongest subject. However, for

us to understand the privilege of self-regulation, I need to explain how we got to this point, so let’s briefly walk through the history… • On February 10, 1899, the NC Bar Association was founded. The goal was to serve lawyers, but also to earn the public’s trust. • In 1903, the NCBA asked the legisla ture to give lawyers responsibility for exam ining, licensing, and disbarring members of the profession. • By 1915, the legislature had authorized a board of legal examiners made up of Supreme Court justices to handle admission and discipline. • By 1932, the members of the NCBA decided that something additional was also needed and began work to establish legisla tion to create the NC State Bar. • On April 3, 1933, the NC State Bar was created by legislation drafted by the NCBA. Self-regulation was not handed to us. We built it ourselves, over decades, and in doing so we made a promise: that we would hold

ning a disciplinary process that is rigorous, fair, and transparent. It should not be a process that shields lawyers from accounta bility; it should be a way that we show the public that the system actually works for them. The Disciplinary Hearing Commission includes eight nonlawyer pub lic members, which is intentional. Our legit imacy depends on public confidence. Self-regulation also shows up in how we handle what has been entrusted to us. The IOLTA program, established in 1983, is a good example. We are required by the Rules of Professional Conduct to hold client funds in trust. When those funds are too small or held too briefly to earn interest for an indi vidual client, the interest those pooled accounts generate goes to civil legal services for North Carolinians who cannot afford a lawyer. The program has put more than $134 million into legal aid since it started. It does not cost taxpayers a dime. It is built entirely on the incidental proceeds of our duty to our clients.

CONTINUED ON PAGE 7

5

THE NORTH CAROLINA STATE BAR JOURNAL

STATE BAR OUTLOOK

Checking Egos at the Door: What North Carolina’s JAG Corps Can Teach the Legal Profession B Y P E T E R G . B O L A C

On an early spring Saturday, I made my way into the Joint Force Headquarters in Raleigh, where the hallways buzzed with the low hum of drill weekend. I had come to meet a group of lawyers whose day jobs couldn’t be more different. It’s easy to imagine North

Colonel Bert Kemp, the Pitt County public defender and the unit’s staff judge advocate, checked in while managing a steady flow of soldiers. And leaders like Colonel (Ret.) Mike McCann, the former staff judge advo

it means there’s a bigger purpose to my work,” says MAJ Jeff Jackson. For MAJ Jefferson Griffin, the gravity of the oath is something he frequently imparts to new attorneys joining the bar. “You’re sit ting here, you’re ready to take an oath...and this is real,” Griffin explains. “And remember every word that you’re saying,” he said, “because this is a heavy profession.” LTC Thorneloe views these oaths as a reflection of character. “I think the oaths that we take, in many ways, just reflect who we already are.” Ain’t No Egos in Chipotle “Maybe it’s because I’ve been here for so long,” MAJ Jackson told me, “but we all just got back from Chipotle, man. Ain’t no egos running wild in Chipotle.” The line stuck with me because it summed up exactly what I had been seeing throughout the day. Once these lawyers put on the uniform, the civilian hierarchy quickly dissolves. None of the political identities, public titles, or par tisan expectations follow them into the guard. Instead, there’s an easy camaraderie and a shared sense of purpose that feels almost rare in their civilian world. “When you put on the uniform, all that stuff fades away,” Jackson says. “There’s never been a moment where any of that par tisanship has intruded on the work that we do together. We are an inherently mission driven and results-oriented organization.” Colonel (Ret.) McCann agrees, highlighting the unique environment the guard provides. “I don’t care what your political position is, and that’s been very good,” he notes, describing the atmosphere as being “colle gial, very family oriented, and bolstered by

cate, and Lieutenant Colonel David Thorneloe, assistant US attorney in the Western District, offered insights dur ing brief lulls in their sched ules. Seeing them this way— rank first, résumé second— underscored how many accomplished North Carolina lawyers quietly ded icate their weekends and annual training to military service.

Carolina’s attorney general, a state appellate judge, a state senator, and a public defend er on opposite sides of a heat ed policy debate, an appellate opinion, or a piece of legisla tion. In the civilian world, the legal profession is inher ently adversarial, and the political climate is undeni ably polarized. Yet one week end a month and two weeks a

year, these exact individuals gather in the same room, put on the same uniform, and work together as a unified team. It’s one thing to read their résumés on paper; it’s another to see these lawyers mov ing through a drill weekend in identical fatigues, slipping in and out of briefings, pausing only long enough to talk between tasks. Over the course of the visit, I met members of the North Carolina National Guard’s JAG Corps wherever their duties allowed. I caught Major Jeff Jackson—the attorney general of North Carolina—in an “office” that genuinely may have been a broom closet. Major Jefferson Griffin, who serves on the North Carolina Court of Appeals, sat down with me in between assignments to share his perspective. Lieutenant Colonel Danny Britt, a state sen ator, and I talked while walking across the compound on his way to the next obligation.

In speaking with these public-servant lawyer-soldiers, a clear theme emerges: the remarkable ability to silence outside noise, set aside personal politics, and subordinate substantial civilian identities to a shared, higher mission. It is a model that the broader legal profession could greatly benefit from emulating. The Power of the Oath Private attorneys are among the few citi zens in North Carolina who take an oath sim ply to perform their daily jobs. For the lawyers in the JAG Corps, that civilian oath (and public official oath) is layered with a military one, creating a powerful foundation of duty. When asked if taking multiple oaths cre ates conflict, the resounding answer is that they reinforce one another. “I love the fact that both of my jobs require an oath, because

SUMMER 2026

6

mutual support.” Colonel Bert Kemp, who serves as the staff judge advocate and manages this group of high-profile leaders, points out that the operational hierarchy is vital to their success. “It’s ironic that you have the AG, or the judge, or a senior senator, all calling me sir!” Colonel Kemp notes, “These individuals are so successful, they could just leave the guard and go do something else...But everybody checks their ego at the door when they get here.” State Senator and Lieutenant Colonel Danny Britt echoes this sentiment regarding their shared meals. “We sit at the table together...and you know, despite our political differences, we’re still brothers in arms.” That quick Chipotle lunch—soldiers shoulder to shoulder, laughing over burritos and swap ping stories—captured the culture better than any formal explanation could. Whatever status they carry Monday through Friday simply doesn’t make it to drill weekend. The lessons these soldiers absorb while wearing the uniform offer more than military discipline; they provide a blueprint for how the legal profession might strengthen its own culture. Servant Leadership LTC David Thorneloe talks about leader ship in terms of shared effort rather than authority. “I don’t ask anybody to do any thing here that I either haven’t done myself or wouldn’t be willing to do with them,” he says. “You put your time in here and you work through those hard things and then you’re ready to lead others.” A Sense of Urgency MAJ Jefferson Griffin sees a clear contrast between military tempo and the pace of civil ian legal work. “I think often, on the civilian side, people get comfortable where they are because it’s not as imminent or it doesn’t feel like we need to move faster,” he notes, advo cating for a greater sense of urgency in the profession. Respect Up and Down the Chain MAJ Jeff Jackson measures leadership by how someone treats the person with the least authority. “You can tell within 90 seconds if this is the type of person who’s going to abuse his power, or whether he’s the type of person who’s going to use it to lift everybody up,” he explains. “The way you tell is whether he e p e . e ” e d y t l g e l t o - r r y Translating Military Values to the Civilian Bar

Left to right: Colonel (ret.) Mike McCann, Lt. Colonel Danny Britt, Major Jefferson Griffin, Colonel Bert Kemp, Lt. Colonel Davis Thornloe, Major Jeff Jackson

treats the lowest ranking soldier with the same respect as the highest-ranking officer.” Civility and Professionalism Colonel Bert Kemp has watched many soldiers set aside their civilian political identi ties the moment the uniform goes on. He believes the profession could benefit from that same discipline. In the guard, he explains, personal philosophies take a back seat to the standards and values of the organ ization. As he put it, “There’s the right way. There’s the wrong way. And there’s the army way.” Our Shared Mission: A Call to Action The North Carolina National Guard’s JAG Corps is a reminder that the legal profes sion, at its best, is an act of public service. Whether they are responding to a natural dis aster in the mountains or advising command ers overseas, these attorneys show how differ ent the work feels when the mission eclipses ego. Politics falls away, personal ambition fades, and the focus shifts to the people who depend on you. The civilian bar has a mission of its own: upholding the rule of law and ensuring justice is done. That responsibility deserves the same

clarity of purpose these soldiers carry into every drill weekend. What I saw in Raleigh were lawyers who checked their titles at the door and put the work first. The legal profes sion could use more clarity. When we treat our shared calling as the priority, not the backdrop, we elevate both the work and the people we serve. n Peter G. Bolac is the executive director of the North Carolina State Bar.

President’s Message (cont.)

Katherine Frye is a founding attorney of Oak City Family Law in Raleigh, where she blends practical strategy with genuine compassion to guide clients through separation, divorce, cus tody, support, and alimony matters. A Campbell Law graduate, she opened her own practice straight out of law school and is now a North Carolina Board Certified Family Law Specialist and a Fellow of the American Academy of Matrimonial Lawyers.

7

THE NORTH CAROLINA STATE BAR JOURNAL

Guarding Against AI Errors: Ethical Risks for NC Attorneys

B Y F R E D W. D E V O R E I I I A N D R O B W I L D E R

B y now, almost everyone has heard of AI, 1 “Artificial Intelligence.” 2 Although some folks refer to it as “almost intelligent” or “artificial ignorance.” AI has moved from theoretical curiosity to an everyday professional tool with aston ishing speed. In the legal profession, AI

now assists with research, drafting, document review, and case evaluation. Used prop

erly, it can enhance efficiency, reduce costs, and improve access to information.

However, when used improperly, AI can undermine professional judgment, violate

ethical duties, mislead courts, and expose attorneys to sanctions and malpractice lia

the-lightwriter/istockphoto.com

bility. When your name is on the signature line, you own every word.

with controlled databases and verifi cation safeguards. Many courts now require you to certify that, in con ducting your research, you either performed the AI searches in the legal databases or that you verified quota tions and citations obtained through AI searches outside of the legal databases. Discussions of “AI in law practice” often treat all uses of artificial intelligence as inter changeable, but that assumption is a mis w r

For those who are new to artificial intelli gence, there are several free websites that will provide “intelligent responses” to queries you propound. Many lawyers use publicly avail able AI platforms such as ChatGPT, Copilot, Gemini, or similar sites. Subscription versions of these sites offer expanded functionality. Meanwhile, major legal research providers—including Westlaw, Lexis, and Bloomberg—have incorporated AI tools into their platforms

take. In practice, lawyers encounter at least three distinct forms of AI. The first is public or general-purpose generative AI, such as ChatGPT, Gemini, or similar tools, which draw from broad, non-curated data sources and present the greatest risk of hallucinated

SUMMER 2026

8

authority. The second is AI embedded with in established legal research platforms, such as Westlaw, Lexis, or Bloomberg, which operate within controlled databases and incorporate citation-checking and verifica tion safeguards—though, as the experiences described below demonstrate, even these tools are not infallible. The third category consists of firm-internal or document review AI tools, often used to summarize discovery, organize records, or analyze large datasets, where confidentiality and supervi sion concerns predominate. Each category presents different benefits and different risks. Treating them as functionally equiva lent invites both ethical missteps and false confidence, particularly when lawyers assume that the presence of a familiar plat form name eliminates the need for inde pendent verification. AI accumulates an almost immeasurable database of books, articles, websites, and conversations, and from this, it learns pat terns of words, facts, and common reason ing. When asked a query, it applies statisti cal analysis and relationships to formulate a response. The problem is that the non-legal databases are not infallible, and AI often makes mistakes. It can provide the attorney with a poignant quotation for a legal issue and provide a citation from which it was derived. But the quotation and citation may be entirely “make-believe.” When this occurs, it is appropriately called a “halluci nation.” AI Hallucinations: Fabricated Authority with Convincing Style One of the most troubling aspects of gen erative AI is its tendency to produce “hallu cinations”—confident, well-structured responses that contain fabricated facts, quo tations, or citations. In legal contexts, hallu cinations often take the form of: • Nonexistent cases • Real cases cited for propositions they never addressed • Quotations that are eloquent, persua sive—and completely fictional. AI does not lie intentionally—but it con fidently generates fiction that looks like law. It generates content that statistically resem bles legal writing. Some people wonder why generative AI hallucinates, and the answer is that it wants to make the user happy. It cre ates content based on its evaluation of the user’s request and its prediction of the best

answer. That "best answer" isn’t always about accuracy; it’s more about coherence and patterns that will keep you engaged. As it learns more, it will become more accurate, but it’s still got a long way to go. Unfortunately, plausibility is not accuracy. 3 When attorneys fail to verify AI-generated output, they risk submitting fiction to the court. 4 The consequences can be severe. Here is what happened to us. Fred’s Story: I represented a client in fed eral court. I prepared a summary judgment brief and consulted with two well-respected attorneys who were experts regarding the legal issues involved. They volunteered to review my brief and make suggestions on its content. The attorneys then collectively sent me several paragraphs of compelling argu ments, complete with persuasive quotations and citations. I gladly incorporated them into my brief. There is no worse feeling than receiving a copy of the defendant’s brief, where the defendant informed the court that the quo tations and citations I had incorporated were hallucinations. I had signed a certification that I did not use AI to write the brief, which was true, but I failed to verify the authentic ity of the supplemental authority I was given. This violated my certification. The fault does not lie with the consulting attorneys because my duty to verify the authorities is non-del egable. My penance was to resubmit a cor rected brief and to warn the Bar of the con sequences of non-approved AI sources. The moral is unavoidable: delegation does not eliminate responsibility. Rob’s Story: I am co-counsel with Fred on the same case in federal court. We agreed to take over the case with very little time left for discovery. We had to review a large vol ume of discovery documents and become deeply familiar with the case and regulatory law in what was, for me, a new area of law. I felt I was behind on technology and hired a consultant to help me get up to date with AI. While writing briefs on dispositive motions, I relied on a commercial legal research pro gram’s rendition of several quotes supporting the points I made in the brief. The cases I cited were correct, but the quotes were hallu cinated. I checked the cases but not the quotes. Again, I was the person who certified that I had authenticated the cited informa tion. Before this incident, I was very confi dent I was using legal AI as effectively as pos sible, only to be shown I was wrong.

Fortunately, the consequences of my mistake have been rectified. Hopefully, my experi ence can help others avoid the same embar rassment, anxiety, and frustration. The Temptation of Over-Reliance on AI Make no mistake about it, AI will forever change how legal drafting and research are conducted. Those attorneys who do not become proficient in the legitimate use of AI will be disadvantaged by opposing counsel who use AI. At the same time, unchecked enthusiasm presents an equal risk. AI’s effi ciency creates a powerful temptation: faster briefs, smoother prose, instant citations. But speed can come at the expense of under standing. When lawyers rely too heavily on AI, they risk substituting output for analysis. In addition to the hallucination issues, AI creates other dangers for the practitioner. Erosion of Professional Judgment Regardless of how efficient AI is and will become, professional judgment cannot be automated. Relying on AI to solve a client’s legal issue is not only a disservice to the client, but it is the path to legal malpractice. Legal judgment is not just correctness; it is also context, nuance, and strategy. AI does not know your judge, jurisdiction quirks, the human dynamics of opposing counsel, or the client’s risk tolerance. Professional judgment develops through reading cases, uncovering weak arguments, and reconciling conflicting authorities. This legal skill will atrophy if there is an over-dependence on AI. Use of AI to draft pleadings or briefs may seem to pro duce polished work quickly. But this is an illusion of mastery and can mask gaps in understanding of the issues and the law. Think of AI as a junior assistant or a second set of eyes, but not as a decision-maker or a proxy for ethical responsibility. Ethical Violations Our State Bar has already adopted formal ethical guidance regarding the use of artificial intelligence in legal practice. 2024 FEO 1 addresses key issues, including when AI assisted legal research is appropriate, whether confidential client information may be uploaded to AI platforms, whether attorneys must disclose their use of AI to clients, the potential impact of AI on hourly billing practices, and the lawyer’s duty to independ ently verify AI-generated research. As the use of AI continues to expand, further updates

9

THE NORTH CAROLINA STATE BAR JOURNAL

being diligent. Efficiency is not diligence if accuracy suffers. Rule 1.6 Confidentiality of Information. “A lawyer shall not reveal information acquired during the professional relationship with a client unless the client gives informed consent.” AI can perform amazing tasks, such as summarizing medical records and providing a timeline, analyzing an expert’s report or deposition, drafting pleadings and discovery materials, and even providing mock jury results (without a mock jury). The problem is that many AI models store the information you upload, making it accessible to others who use the AI. The approved legal databases should provide you with an “infor mation silo,” which means that your client’s information is protected and segregated from others accessing the database. Rule 2.1 Advisor. “In representing a client, a lawyer shall exercise independent, professional judgment and render candid advice. In rendering advice, a lawyer may refer not only to law, but also to other con siderations such as moral, economic, social, and political factors that may be relevant to

the client’s situation.” This Rule takes on new meaning with the proliferation of AI. Independent, professional advice does not mean “AI said it, so it must be true.” Rule 3.1 Meritorious Claims and Contentions. “A lawyer shall not bring or defend a proceeding, or assert or controvert an issue therein, unless there is a basis in law and fact for doing so that is not frivolous, which includes a good faith argument for an extension, modification, or reversal of exist ing law...” More than once have I submitted a query to AI whose response, upon verifica tion, was false. The lawyer cannot accept AI summaries or legal conclusions without undergoing independent scrutiny. Otherwise, the attorney runs the risk of advancing non-meritorious claims—even unintentionally. Rule 3.3 Candor Toward the Tribunal. “(a) A lawyer shall not knowingly: (1) make a false statement of material fact or law to a tribunal or fail to correct a false statement of material fact or law previously made to the tribunal by the lawyer...” Courts rely on attorneys to verify their authorities. Judges

and revisions to this opinion are inevitable. And, while AI will most likely prompt changes to our Rules of Professional Conduct, our existing ethical rules already warn against AI’s misuse. 5 Rule 1.1 Competence. “A lawyer shall not handle a legal matter that the lawyer knows or should know he or she is not com petent to handle without associating with a lawyer who is competent to handle the mat ter. Competent representation requires the legal knowledge, skill, thoroughness, and preparation reasonably necessary for the rep resentation.” AI can create a false sense of competence, encouraging lawyers to take on matters beyond their experience because the tool appears to “know” the law. Apparent fluency does not equal actual understanding. Rule 1.3 Diligence. “A lawyer shall act with reasonable diligence and promptness in representing a client.” AI may encourage speed over independent judgment. It may encourage a cursory review of the legal issue rather than a deep analysis. The tendency would be for the lawyer to become “lazy,” and when it occurs, the attorney is not

SUMMER 2026

10

are not fact-checkers for lawyers’ citations. Submitting false authority violates this duty, regardless of intent. In our confessions above, neither of us intended to mislead the court about our legal authorities. But the burden of complying with Rule 3.3 lies exclusively with the attorney. Rule 5.3 Responsibilities Regarding Nonlawyer Assistance. With respect to a nonlawyer employed or retained by or asso ciated with a lawyer: “(a) a principal, and a lawyer who individually or together with other lawyers possesses comparable manage rial authority in a law firm or organization, shall make reasonable efforts to ensure that the firm or organization has in effect meas ures giving reasonable assurance that the nonlawyer’s conduct is compatible with the professional obligations of the lawyer...” While this Rule was written prior to the invention of AI, it carries an interesting new perspective if AI is considered a non-lawyer assisting in the representation of a client. AI also complicates traditional supervision in ways that lawyers may not immediately appreciate. Junior lawyers may assume that AI-generated quotations embedded in com mercial research platforms have been verified. Supervising attorneys may assume that jun iors independently checked the underlying sources. In reality, AI can sever the usual chain of responsibility by obscuring who actually verified what. When an error sur faces, courts will not be interested in whether the mistake originated with software, a con sultant, or a junior lawyer. They will focus on the attorney whose name appears on the fil ing. AI does not change the supervisory obli gations imposed by the Rules of Professional Conduct; it makes them more demanding. Firms that fail to train lawyers—especially younger ones—on verification and disclosure risks are inviting problems that no amount of post hoc explanation can fix. Rule 8.4 Misconduct. It is professional misconduct for a lawyer to: “(a) violate or attempt to violate the Rules of Professional Conduct, knowingly assist or induce another to do so, or do so through the acts of anoth er;...(c) engage in conduct involving dishon esty, fraud, deceit or misrepresentation that reflects adversely on the lawyer’s fitness as a lawyer; (d) engage in conduct that is prejudi cial to the administration of justice;...(g) intentionally prejudice or damage his or her client during the course of the professional relationship, except as may be required by

Rule 3.3.” Submitting false, misleading, or unverified information—whether knowingly or recklessly—can prejudice the administra tion of justice and reflect adversely on fitness to practice law. Rule 11: Both the North Carolina and federal rules of civil procedure provide that an attorney may be sanctioned for making representations to the court that have not been verified to the best of the attorney’s ability. We don’t see relying on AI to be an exception to the Rule. Sanctions—Courts are No Longer Patient Every lawyer who uses AI should read the 51-page sanction order in Johnson v. Dunn , 792 F. Supp. 3d 1241, 1246 (N.D. Ala. 2025) (Judge Anna M. Manasco). In that case, three attorneys for the defendant filed a brief in support of a motion to compel. One of the attorneys drafted the brief and unknowingly incorporated five AI halluci nations from ChatGPT. His co-counsel signed the pleading without verifying the quotations and citations. In the order, the judge discussed Rule 11 sanctions and viola tions of the Rules of Professional Conduct. After conducting a show cause hearing, Judge Manasco entered an order with the following sanctions: (1) the attorneys were disqualified from the case; (2) a copy of the order had to be provided to their clients, opposing counsel, and presiding judge in every pending state or federal case in which they were attorneys of record; (3) the clerk of court was directed to submit the order for publication in the Federal Supplement; (4) the offending attorneys were ordered to pro vide to the clerk of court a listing of every jurisdiction in which they were licensed to practice law within 24 hours of the entry of the order; and (5) The clerk was directed to serve a copy of the order on the general counsel of the Alabama State Bar and any other licensing authority. This order should frighten all of us. In her opinion, the judge wrote: “Every lawyer knows that citing fake cases in a court filing is a terrible decision. No one here is attempting to defend it. In the few years that generative AI has affected court filings, it has become well established that “[m]any harms flow from the submission of fake opinions.” Id. at 1256. Legal Guardrails for the Use of AI The key here is that AI can assist but

never replace independent professional judg ment. 1. No AI-generated citation may be relied upon unless located in a trusted legal data base (Westlaw, Lexis, Bloomberg, or other official court-approved sites). 2. Any non-legal database research must be carefully scrutinized and verified through an approved legal database. 3. When using AI to review a client’s file (i.e., summarize medical records), never input client names, confidential facts, privi leged documents, settlement positions, or personally identifiable information unless the AI database is explicitly approved for confidential use. You must be sure that, even in approved legal databases, any information you enter is kept in a confidential knowledge silo that cannot be accessed by others. Unless you have purchased a subscription from a non-legal database that offers unquestioned confidentiality, then any information or doc ument you upload in a query could become part of the sea of knowledge AI might use in a similar query from another user. 4. Firms must set clear usage rules. Just as when a partner relies on a junior lawyer’s work, the responsibility rests at the feet of the filing attorney. “The software did it” is not a defense. 5. Each firm needs to develop a policy regarding the use of AI. It needs to incorpo rate human oversight and quality control. There should be training on security and confidentiality concerns. Before You File an AI-Assisted Brief Before submitting any pleading, motion, or memorandum that involves AI at any stage, ask yourself: ☐ Have all citations been independently located in a trusted legal database (Westlaw, Lexis, Bloomberg, or an official court source)? ☐ Have all quotations been checked against the original source, not merely accepted from AI-generated summaries? ☐ Can I explain, if asked by the court, exactly how AI was used and what steps I took to verify its output? ☐ Did I avoid inputting confidential, privileged, or identifying client information into any AI platform not expressly approved for confidential use? ☐ Would I be comfortable defending my use of AI at a show cause hearing, under oath?

11

THE NORTH CAROLINA STATE BAR JOURNAL

except in approved legal databases. No client information should ever be uploaded to an AI model unless there is verification that the information will remain confidential and will be accessed only by the firm. Attorneys remain fully responsible for all work product and must comply with all confidentiality, competence, and candor obligations under the Rules of Professional Conduct. Conclusion AI will continue to revolutionize legal practice. Its benefits are substantial. But innovation does not diminish responsibility. The law demands accuracy, judgment, and integrity—qualities that cannot be automat ed. Used wisely, AI enhances human capabil ity. Used carelessly, it undermines the profes sion. The lesson is simple, but unforgiving: AI may draft it, but when you sign it, it’s yours. Someone once said, “Make AI a tool for humans, not a reason to lose what makes us human.” n Fred DeVore is a partner in DAS Law Group. He has been a trial attorney for over 43 years. Fred was formerly the president of the Mecklenburg County Bar. He is currently a State Bar counselor and has previously served on the State Bar’s Disciplinary Hearing Commission. He has been a frequent speaker at CLEs on the use of technology in the courtroom. Rob Wilder has been a trial lawyer in Charlotte for over 50 years. He concentrates his practice on business disputes and complex civil litigation in state and federal courts. Endnotes 1. Sometimes AI is referred to as LLM (large language model). 2. However, on April 8, 2025, the US Secretary of Education, Linda McMahon, described using “A1” in the courtroom. Presumably, she was referred to the technology and not the steak sauce. 3. An interesting article on this topic is “Plausibility is Not Truth, Do You Really Understand AI?” by Robert Hutton, thecritic.co.uk/plausibility-is-not-truth. 4. Gina Esfandi in Advocate has published an illuminating article entitled, “The Potential and Drawbacks of Using Artificial Intelligence in the Legal Field,” bit.ly/4cbjwoK. 5. The American Bar Association has published an excel lent article entitled “Top Six Legal Issues and Concerns for Legal Practitioners.” bit.ly/4ccrwWA. 6. This sample policy has not been approved by any legal body. You should rely on your own research in judg ment when finalizing a policy for your firm.

provide guidelines for your firm. 6

☐ If a junior lawyer, consultant, or ven dor used AI, did I personally review and ver ify the resulting work product? ☐ If required by local rule or standing order, have I complied with all AI disclosure or certification requirements? If the answer to any of these questions is “no,” the filing is not ready. Sample Firm Policy (Simplified) Here is a draft of a firm policy that may

Attorneys may use approved AI tools as drafting and research aids, provided that all outputs are independently reviewed, verified, and approved by a licensed attorney. Provided, however, that the use of AI platforms may be limited by court requirements, including standing orders and local rules. AI tools may not be relied upon for legal authority, strate gic decisions, or ethical determinations

SUMMER 2026

12

13

THE NORTH CAROLINA STATE BAR JOURNAL

Beyond the Ban: Why Your Law Firm Needs a Realistic AI Policy in2026

B Y C A T H E R I N E R E A C H

I n 2026, artificial intelligence is deeply embed ded in legal and business operations, making clear policies essential. Simply telling lawyers not to use AI is unrealistic, as these tools are now part of everyday technology. Without defined guidelines, law firms risk confiden tiality breaches, ethical missteps, and could lose client trust. A robust AI use policy

is vital for balancing innovation with responsibility. The good news is that there

is guidance out there, so you don’t have to start with a blank slate.

DrAfter123/istockphoto.com

By 2025, the legal industry had moved past the question of if artificial intelligence will be used, to how it must be governed. According to the Clio Legal Trends report, 79% of legal professionals utilized AI tools, but 44% of law firms had yet to implement formal governance policies. Why “Just Say No” Doesn’t Work In an effort to mitigate risk, some firms have imposed blanket bans on the use of gen

erative AI. While well-intentioned, these restrictive policies are often unrealistic and counterproductive for several reasons: The “Shadow AI” Risk

When firms ban AI without providing approved alternatives, they inadvertently cre ate “Shadow AI,” which involves the unau thorized use of tools by employees without IT knowledge. Lawyers, under pressure to be efficient, may turn to free, consumer-grade

tools (like the free version of ChatGPT) on personal devices to draft emails or summarize documents. This is far riskier than controlled adoption because the firm loses all visibility into where client data is going, and con sumer tools often use inputs to train their

SUMMER 2026

14

models, leading to potential confidentiality breaches. AI is Already Everywhere A total ban is practically impossible to enforce because AI is no longer just a stand alone chatbot; it is embedded in the soft ware lawyers use daily. From Westlaw and Lexis+ to Microsoft 365 and Zoom, AI fea tures are integrated into the very infrastruc ture of modern legal practice. Blocking AI entirely would effectively mean blocking the industry’s standard operating tools. The Efficiency Imperative Clients expect efficiency. If a firm charges for five hours of work that could be done in one hour with AI, they risk being undercut by competitors who have safely integrated these tools. As one source notes, prohibition drives usage underground; clear policies bring it into the open where it can be supervised. Instead of a blockade, firms need a guardrails policy that empowers lawyers to use technology safely while strictly adhering to ethical and legal obligations. In North Carolina, law firms can use 2024 Formal Ethics Opinion 1 “Use of Artificial Intelligence in a Law Practice” to help guide policy. Other helpful guidance can be found in the ABA Task Force on Law and Artificial Intelligence report ( see bit.ly/4mqQoNw), as well as other state ethics opinions and task force reports ( see bit.ly/3Q6OBBj). Here are some components of an effec tive AI use policy: Establish a Risk-Based “Traffic Light” System In their step-by-step guide for crafting an AI policy for your law firm, legal AI company Casemark suggests a classification system to streamline approval ( see, bit.ly/4dNuTnT): · Red Light (Prohibited): Inputting con fidential client data into public/con sumer AI tools; using AI for fact-finding without verification; automated deci sion-making for client outcomes. · Yellow Light (Oversight Required): Legal research, document review, and first drafts. These require adherence to specific verification protocols. · Green Light (Standard Use): Administrative tasks, marketing content, Building a Thoughtful Policy: Best Practices

and internal scheduling. The “Human in the Loop” is Non Negotiable The policy should explicitly state that AI is a tool, not a replacement for professional judgment. Due to the persistent risk of hal lucinations, policies should mandate that every AI-generated output be verified. This includes independently confirming case citations and validating factual assertions against source documents. Transparency and Client Consent Policies should address when and how to inform clients about AI use. This includes updating engagement letters to disclose that the firm uses AI tools to enhance efficiency while maintaining human oversight and con fidentiality. The Illinois Attorney Registra tion and Disciplinary Commission’s Guide to Implementing AI has sample language for client notification and consent. See bit.ly/4260BFL. Continuous Education, Not Just Rules A policy document is useless if no one understands it. Firms must pair their poli cies with mandatory training. Training will not only help reinforce the specifics of the policy, but also provide an opportunity to review appropriate tools, share best prac tices for the use of generative AI, and dis cuss ways the firm can increase efficiency withAI. Sample Law Firm AI Policies When preparing to develop your firm’s AI policy, you will find numerous model policies, sample documents, and guidance materials available to assist you. While these resources offer valuable starting points, it is important to recognize that they may not fully address the unique circumstances of every law firm. They are designed to serve as preliminary frameworks rather than compre hensive solutions. • Implementing an AI Policy: What to Know (Brent J. Hoeft, State Bar of Wisconsin, bit.ly/4teMqua). • Law Firm Acceptable Use Policy for Artificial Intelligence (AI) and AI Policy and Governance Guidance (Texas Bar Practice, bit.ly/422wRcW) • Sample Law Firm GAI Use Guideline/Guidance Policy (American Inns of Court, bit.ly/3QoLKDP) • VBA Model AI Policy for Law Firms (Virginia Bar Association, vba.org/?pg=ai) • Sample Use of GAI Tools Policy (Illi

nois Attorney Registration and Discipli nary Commission, bit.ly/4t4l7m6) • How to Write a Simple Firm Policy on the Acceptable Use of Generative AI for Client Work (LeanLaw, bit.ly/4tdne7j) • Sample Generative Artificial Intelligence Policy (Centerbase, bit.ly/3QEzT4z) • Crafting an AI policy for your law firm: a step-by-step guide (2025 Edition) (Casemark, bit.ly/4syP24K) • Responsible AI Use Policy Outline for Law Firms (JusticeatWork, bit.ly/4cND3L5) • AI Policy Template for Law Firms (Plus Examples) (Darrow.ai, bit.ly/3QmIYim) • Crafting an AI Use Policy: The Essential First Step to Unleashing AI in Your Law Firm (Eve.legal, bit.ly/3Q7MgpM) • Model Artificial Intelligence (AI) Policy for Law Firms (Allrize, bit.ly/4mqAmTS) • Law Firm AI Policy Template, Tips & Examples (Clio, bit.ly/3QmJih4) Conclusion This gap between adoption and over sight exposes firms to significant risks, from data breaches to court sanctions. However, the reaction to banning the use of AI entire ly in a law firm is unrealistic. New lawyers in your firm are receiving training in law school on how to use AI. AI is embedded in most personal and business technology from smartphones to operating systems to search engines to office suites. Some clients may even question whether a firm is lever aging AI tools for efficiency. By implement ing a policy that acknowledges the reality of AI usage while enforcing strict ethical boundaries, firms can turn a potential lia bility into a competitive advantage. n Catherine Reach is the director of the North Carolina Bar Association’s Center for Practice Management. This article, Beyond the Ban: Why Your Law Firm Needs a Realistic AI Policy in 2026, originally appeared on the From the Center blog, published by the North Carolina Bar Association (bit.ly/428fs2u). It is repro duced here with the permission of the author, Catherine Sanders Reach, and the copyright holder, North Carolina Bar Association. © 2026 North Carolina Bar Association. All rights reserved. Any further reproduction or distribution of this material is prohibited without the express written permission of the North Carolina Bar Association.

15

THE NORTH CAROLINA STATE BAR JOURNAL