Camille Stell is Vice President of Risk & Practice Management for Lawyers Mutual Liability Insurance Co. of NC. Continue this conversation by contacting Camille at camille@lawyersmutualnc.com
For the past few years, when I talked with lawyers about artificial intelligence, the conversation usually started with some version of the same question, “Should lawyers really be using this?”
In 2026, that conversation has changed.
Increasingly, lawyers aren’t asking whether they should use AI. They are already using it. Now they are asking better – and harder – questions: What information can I safely put into an AI tool? How do I know whether a product is secure? Can I trust the research it produces? Do I need an AI policy for my firm? Should I tell my clients I’m using it? And if AI lets me complete in 20 minutes something that once took two hours, how do I bill for that work?
In other words, AI has moved from being an interesting technology topic to being a law practice management issue.
Lawyers Are Already Using It
The numbers tell part of the story. Clio’s 2026 research reports that 71% of solo practitioners and 75% of small firms are now using AI to complete legal work. Smaller firms tend to use AI more selectively than larger organizations, often relying on general-purpose products such as ChatGPT or AI-enabled legal research tools rather than implementing AI throughout the entire firm.
That makes sense. A solo or small firm lawyer often wears many hats. On any given day, you may be lawyer, researcher, writer, marketer, HR director, intake coordinator and office manager. AI can help with many of those functions from summarizing and drafting to client communications, scheduling, intake and administrative work.
We are seeing something similar among North Carolina lawyers.
During an AI webinar Lawyers Mutual hosted this spring, we polled 200 insured lawyers participating in the program. About 70% reported using AI in their legal work at least occasionally, while roughly 30% said they weren’t using it at all. ChatGPT was by far the most commonly identified tool, followed by Microsoft Copilot, with lawyers also reporting use of Gemini, Claude, Clio, Lexis+ AI and other products.
Keep in mind, this was not a scientific survey. These lawyers had voluntarily signed up for an AI webinar, so they were presumably more interested in AI than the average lawyer. Still, the results were remarkably consistent with the national data.
And the interest is unmistakable. Our typical Lawyers Mutual webinars averaged about 40 attendees. Our AI webinars this year have averaged approximately 200 attendees. Of the 50 live CLE programs we offered in 2026, more than one-third were on the topic of AI.
The Bigger Concern: Adoption Is Outpacing Governance
Here is where I think law firms need to pay particular attention.
Lawyers are experimenting with AI much faster than many firms are developing rules for using it.
Our webinar poll found that many respondents either had no formal AI policy or were operating with informal guidance. Few had changed their billing policies in response to AI, and only a minority reported advising clients about the risks associated with entering confidential information into an AI platform.
That gap between AI adoption and AI governance may be one of the most important risk-management issues facing firms right now.
The North Carolina State Bar has already given lawyers a framework. In 2024 Formal Ethics Opinion 1, the State Bar made clear that lawyers may use generative AI. But our traditional professional responsibilities don’t disappear simply because technology is doing some of the work.
We still have obligations involving competence and confidentiality. We still have to supervise work performed on behalf of our clients. And we still have to independently review the work product before relying on it.
That means the question shouldn’t simply be, “Can I use ChatGPT for this?”
A better set of questions might be: What happens to the information I enter? Does the vendor retain it? Is it used to train the model? Have I verified the output? Does someone else in my office know enough to recognize when AI has produced a convincing but incorrect answer?
And perhaps most importantly – Am I still exercising my own professional judgment?
Start With the Problem, Not the AI
One of the mistakes I think firms can make is beginning with the technology – We need to start using AI. What should we buy?
Try turning that around.
Instead ask what problem in our firm are we trying to solve?
Maybe lawyers are spending too much time drafting routine correspondence. Maybe intake is inefficient. Perhaps meetings aren’t being documented consistently. Maybe lawyers struggle to turn their notes into follow-up tasks. Or perhaps you want a better first draft of marketing content, policies or internal communications.
Those are good places to explore whether AI can improve an existing workflow.
At Lawyers Mutual, that thinking contributed to our first partnership with an AI company, Querious AI. Querious is a confidential AI-powered note-taking tool designed specifically for lawyers. Instead of simply generating legal work product, it addresses a familiar law-practice problem – documenting client conversations and identifying the tasks that need to happen afterward.
That is the kind of AI use I find particularly interesting from a risk-management perspective. Many malpractice problems don’t begin because a lawyer doesn’t know the law. They begin with a communication breakdown, inadequate documentation, or something that didn’t get followed up on.
AI that helps us build better processes around those problems may ultimately be just as valuable as AI that can draft a brief.
AI Is Changing Client Expectations
There is another change coming that lawyers shouldn’t overlook.
If AI allows us to accomplish in minutes what once took hours, clients are eventually going to ask why they should continue paying for that work in the same way.
According to Thomson Reuters January 2026 Future of Professionals Report, 71% of in-house legal professionals expect outside law firms to change their commercial models as AI use increases, while only 28% of law firms reported changing their pricing structures in response to AI.
For solo and small firms, that could create opportunities as well as challenges. Greater efficiency may make flat fees, value pricing and other alternative fee arrangements more attractive. A small firm that thoughtfully incorporates AI may also be able to handle more work without immediately adding personnel.
But efficiency doesn’t eliminate the lawyer’s value. If anything, AI may make judgment, experience, strategy, empathy and trusted advice even more important.
Clients aren’t simply paying us to produce words on a page. They are paying us to understand their problem and help them find a solution.
And AI Isn’t Just Changing Lawyers
This summer while attending the various bar association summer conferences, an AI development many conference attendees were discussing is the use of AI from self-represented litigants.
Generative AI can produce something that looks remarkably like a legal pleading, motion or brief, even when the document contains nonexistent cases, incorrect legal propositions or arguments the person submitting it doesn’t fully understand.
This creates a tension that lawyers and judges are forced to deal with. AI potentially improves access to legal information and helps people articulate legitimate claims. At the same time, it can dramatically increase the volume and apparent sophistication of material that courts and opposing lawyers must evaluate.
Courts are beginning to respond. Federal and state courts in North Carolina have started adopting orders addressing AI-assisted research and drafting while emphasizing that the person submitting the document remains responsible for its factual and legal accuracy.
This is another reason every lawyer needs at least a working understanding of generative AI, even a lawyer who has personally decided not to use it.
Your clients may be using it. Your employees may be using it. Opposing parties may be using it. And the person who walks into your office for a consultation may arrive with 30 pages of AI-generated “legal research” already in hand. Or they may plan to leave your office and feed their privileged, confidential work product into an AI tool for a second opinion.
Where Do We Go From Here?
I don’t think the most useful advice for lawyers is simply “use AI” or “don’t use AI.”
Instead, we need to become competent, informed and disciplined users of AI.
This means understanding confidentiality and cybersecurity. It means vetting vendors rather than simply signing up for the latest tool. It means having policies for employees and training them on those policies. It means verifying AI-generated work rather than assuming that a polished answer is a correct answer. It means thinking about client communication, billing and engagement letters. And throughout all of it, it means preserving independent professional judgment.
The AI conversation is going to continue to move quickly. The specific products we are talking about today may not even be the products we’re talking about two years from now.
But the fundamental risk-management principles aren’t new at all.
Protect your client’s information. Know the tools you use. Supervise the people, and technology, working on your client’s matter. Check your work. Communicate with your client. Charge reasonable fees. And exercise professional judgment.
AI may be changing how we practice law.
It hasn’t changed what it means to be a lawyer.
Camille Stell is the Vice President of Risk & Practice Management for Lawyers Mutual and the co-author of the book, RESPECT – An Insight to Attorney Compensation Plans available from Amazon. Continue this conversation by contacting Camille at camille@lawyersmutualnc.com or 800.662.8843.
