Someone at your firm has already wondered this. Maybe not out loud, and maybe not to you.
They were staring down a client’s trial balance, or a stack of statements that needed summarizing, or a memo that would take ninety minutes to write and twelve minutes to write with a little help. And they wondered — briefly, and then not so briefly — whether they could just paste it in.
Some of them wondered and stopped. Some of them didn’t.
You need a real answer for this. Not a hedge, not “we’re looking into it,” and not a blanket no that everyone quietly ignores anyway. Here’s the honest one.
The answer depends entirely on which door the data walks through
“Can we use AI with client data” isn’t really one question. It’s a question about where the tool sits.
A personal ChatGPT account is a different product than a company-managed tenant. Same model, same friendly interface — a materially different agreement underneath. When someone signs up with a personal email and a personal card, they’ve quietly entered your firm into a consumer arrangement that nobody reviewed. Data handling, retention, whether inputs help train the model, what happens if that account gets compromised — all of it is governed by terms nobody at your firm has read, and none of it is something you control.
A licensed business tenant is a different animal entirely. Data handling is contractual. Retention is defined. You own the tenant, you control who’s in it, you can shut it off, and — this is the part that actually matters — you can describe it to a client who asks. If your firm runs on Microsoft 365, you may be closer to a defensible answer than you think. Often the tool is already sitting inside the license you’re already paying for.
So the honest answer: it depends on whether your people are working inside something your firm controls, or outside it. And right now, at most firms, the truth is some of both — and nobody knows the split.
The three questions your staff cannot currently answer
Not because they’re careless. Because nobody has told them.
“Is this tool approved?” If your firm has never named one, every person is making that call on their own, every week. They’ll keep making it, and they’ll keep making it differently.
“Is this specific piece of information okay to put in?” There’s a real difference between a redacted excerpt, a public filing, and a client’s full general ledger. Your staff knows that instinctively. What they don’t have is a line — an actual written line — telling them where it sits. Without that line, they guess, usually under deadline pressure, usually alone.
“Who do I ask when I’m not sure?” This is the one that quietly matters most. If the honest answer is “nobody,” then the default behavior when someone’s unsure is to proceed and say nothing. That’s the exact situation your firm can’t afford.
Every one of these questions is answerable in a single sentence. None of them are answered at your firm today.
The reflex to ban it is the expensive one
Confidentiality isn’t negotiable in this business. Neither is the human relationship your clients pay for — nobody wants to feel like their return got run through a machine.
But an outright ban protects neither of those things. It just moves the behavior to phones and home laptops, where you have zero visibility. And it hands a real advantage to the firm down the road that took the time to do this properly.
Because they are doing it properly — and here’s what it’s returning: hours back on document review, on research, on first-draft memos, on the reconciliation work that eats a senior associate’s whole afternoon and generates zero client value. That’s billable capacity you’re currently spending on tasks a machine could carry, right in the middle of busy season, when you can’t exactly hire your way out of it.
The firms getting this right didn’t loosen their standards on confidentiality. They put a clear boundary around it, and then they moved.
What you actually need in writing
Less than you’d think. This isn’t a compliance program.
- One approved tool, named. Ideally the one already sitting inside the license you’re already paying for.
- One page that spells out what client information may go in, what may never go in, and what needs a person’s eyes on it before it leaves the firm. Plain language, not legal language — because it has to get read on a busy Tuesday, not filed away.
- One name — the person to ask when the answer isn’t obvious. So “I wasn’t sure” ends in a question, not a shortcut.
- One sentence you can say to a client who asks about your firm’s position on AI. You will be asked. Better to have the answer ready before it happens.
That’s the whole document. It takes an afternoon. And the day it exists, your firm stops being a place where people are guessing quietly on their own, and starts being a firm where people are working inside a boundary you set on purpose.
We’ve spent nearly three decades keeping Savannah-area businesses — from law firms to medical practices to firms handling other people’s money — running on technology they can actually trust and explain. This is the same instinct, applied to the newest tool on the desk.
Get the AI Acceptable Use Policy Starter Kit
A one-page, plain-language policy your team will actually read — plus the sanctioned-tool checklist, the data-classification lines to draw, and the answer to give a client who asks.
Written for firms handling confidential client information. Free.