When Clients Ask About Your AI Strategy (Here's What to Say)

When Clients Ask About Your AI Strategy (Here's What to Say)

The question came up in a client review meeting last month. A general counsel at a mid-size company asked the managing partner of a 14-lawyer employment firm: "What are you doing with AI?"

The managing partner hadn't prepared an answer. She gave a vague response about "exploring tools." The client nodded and moved on. The firm kept the account — for now.

That moment is happening at firms like yours all over the country. The difference between firms that handle it well and firms that stumble is not AI sophistication. It's preparation.

Law firm clients asking about AI is no longer a hypothetical. It's a weekly occurrence. And for small firms — where one uncomfortable client conversation can cost a year of revenue — having a clear, confident answer isn't optional.

---

The Clients Are Already Using It

Here's why this question is coming from clients now, in 2026, instead of two years ago: they've crossed.

According to a Q1 2026 survey by the ACC and Everlaw, 87% of general counsel are now using generative AI within their own legal teams — nearly double the 44% adoption rate from just one year ago. These are not early adopters. These are in-house lawyers who adopted AI because they had to, experimented with it, and are now evaluating every vendor relationship through a new lens.

When your client asks "what are you doing with AI?", they're not asking because they read a trend piece. They're asking because they made the crossing themselves. And now they want to know whether you have too.

That shifts the conversation. This isn't a client grilling you about technology they don't understand. It's a client who uses AI daily asking whether their outside counsel is keeping pace.

The FTI Consulting/Relativity research adds another layer: 64% of in-house counsel expect generative AI to reduce their reliance on outside counsel over the next two to three years. They don't need as much external help with routine work. The firms they'll still call are the ones who can offer faster turnaround, sharper analysis, and transparent use of the same tools the client is already using.

Being asked about your AI strategy is not a threat. It's a business development moment.

---

What the Litera Survey Actually Found

In May 2026, Litera — a legal technology company — published its State of Legal AI: Spring 2026 Market Sentiment Report. The headline numbers are worth knowing before your next client meeting.

85% of law firms surveyed are already experiencing or expecting direct client pressure on their AI strategy. This is not a future trend. It's current reality across firms of all sizes.

51% of respondents said a client directly influenced an AI investment decision in the past 12 months. Clients aren't just asking — they're changing what firms buy and how firms work.

Only 18% of law firms track AI return on investment. Here's the problem: clients are asking for accountability that most firms aren't even measuring for themselves.

Only 15% describe their AI investment as still entirely internally driven. Everything else is influenced, at least in part, by client expectations.

The survey was conducted across firms of all sizes, which means the headline numbers skew toward larger AmLaw firms with more resources and more enterprise clients. But the pattern hits boutiques and small firms on a delay — and then it hits harder, because small firms have fewer staff, thinner margins, and less room to absorb client attrition.

If you run a 5–20 lawyer firm and you haven't been asked this question yet, you will be.

---

The Three Questions Clients Are Actually Asking

Most "AI strategy" questions from clients reduce to three things. Know the answer to each before you're in the room.

1. "Are you using AI on my work?" (The Disclosure Question)

This is the most common version of the question. It sounds simple but it carries real weight — your client is asking whether your deliverables are AI-generated, AI-assisted, or human-only.

Your answer framework: Be specific about what you use AI for, and honest about what you don't. "We use [tool] to accelerate first-draft contract review and research. Every AI-generated output is reviewed and edited by a licensed attorney before it reaches you." If you don't use AI yet, say that clearly — and briefly note that you're evaluating tools. What you don't want to do is give a vague non-answer that sounds like evasion.

2. "How does AI affect what you charge me?" (The Billing Question)

This question is the one that makes firm owners most uncomfortable, because the honest answer is complicated. AI does reduce time on certain tasks. Whether that translates to lower invoices or to better work at the same price is a firm-by-firm decision — but clients are entitled to understand your position.

Your answer framework: "AI tools have made certain tasks faster. We're using those efficiency gains to [deliver faster turnaround / allocate more attorney time to complex analysis / keep prices stable despite inflation]. Here's how that shows up in your matters: [specific example]." You don't have to discount. But you do have to have a coherent answer.

3. "Is my information safe?" (The Confidentiality Question)

This is the question that gets asked by the most cautious clients — usually after they've read something alarming about AI data practices. It's also the easiest to answer well if you've done the work.

Your answer framework: "We only use enterprise-licensed tools with data processing agreements that prohibit training on submitted content. We never input client names, matter numbers, or personally identifying information into consumer-tier AI products. Our AI usage policy is available on request." If you haven't yet evaluated your tools on this dimension, reviewing your AI stack is the first thing to do before this conversation happens.

---

Building Your One-Paragraph AI Answer

Firms that handle the client AI question well have one thing in common: they prepared a short, clear answer in advance and they use it consistently. Here's a template you can adapt.

"Our firm uses [tool or category — e.g., a legal AI research assistant] for [task — e.g., initial case research and contract review]. It saves approximately [X hours] per matter, which we [pass on to clients as faster turnaround / reinvest in deeper attorney review / use to keep our pricing competitive]. Every AI-assisted work product is reviewed by a licensed attorney before delivery. We have a written AI usage policy that covers data handling and confidentiality — I'm happy to share it."

Walk through each component and make it yours:

• Tool or category: Name the tool if it's well-known (Harvey, Clio's AI features, Westlaw Precision). Use a category if you'd rather not be specific ("a legal research AI tool with enterprise data protection").
• Task: Be specific. "Research" is too vague. "First-pass contract review for vendor agreements" is specific enough.
• Time savings: If you've tracked it, use the actual number. If not, "significant" is acceptable — but a number builds more trust.
• What you do with the savings: This is where you differentiate. Some firms pass it on as price stability. Some use it to give clients faster turnarounds. Some invest it in attorney capacity. All are defensible — pick yours.
• Attorney review: Non-negotiable. This line protects you ethically and reassures the client.
• Policy available: This alone — the fact that you have a policy — separates you from most small firms.

Put a version of this answer in your engagement letter. Clients who sign an engagement letter with a clear AI disclosure clause don't ask this question in year two of the relationship. They already know.

---

When the Client Asks Before You're Ready

You may be reading this after the question already came up and you gave a fumbling answer. That's fine.

If a client asks about your AI strategy and you don't have a polished answer yet, here's the honest pivot:

"We're in the process of formalizing our AI approach and I want to be able to give you a complete answer. Can I follow up with you by [specific date — no more than two weeks]? In the meantime, here's how we currently handle [the specific concern they raised]."

This is better than bluffing. Clients generally respect "I'll get back to you with something specific" far more than a vague answer that sounds like you're hiding something.

Then actually follow up. Set a calendar reminder. When you do, send your one-paragraph answer and a brief note that you've put it in your engagement letter going forward.

That follow-up — done well — is better for the relationship than having a ready answer in the first meeting. It shows you took their question seriously.

---

The Moment in Front of You

You probably won't have to wait long for the next client AI question. The Litera data says 85% of firms are already in this conversation.

Here's the reframe: every client who asks about your AI strategy is giving you an opening. They're asking because they care about how you work. They're asking because they're evaluating you. A firm that answers confidently, with a specific tool, a clear policy, and a sentence about what they do with the time savings — that firm looks forward-thinking.

A firm that fumbles it, or deflects, or says "we're looking into it" — that firm looks like it's behind.

You don't need to be running an AI lab. You need a one-paragraph answer and a policy you can share. That's it.

---

Want help building the AI policy you reference in that conversation? Start with our Law Firm AI Policy Template — a one-page, copyable document built for firms with 5–25 lawyers.

For guidance on what to put in your engagement letter, see our post on AI disclosure language for engagement letters.

Get the weekly Crossing Report — practical AI intelligence for professional services firm owners — delivered every Monday at 6 AM.