Need a way to answer questions and secure leads at 2:37 in the morning? A law firm chatbot could be exactly what you need, with a few caveats. Most firms use these convenience tools to help filter potential clients and answer some common questions, but without the right ethical safeguards in place, those same firms can find themselves staring down the barrel of bar complaints.
Generative AI is changing the way every industry does things, and the rules around using it need to keep up with the times In July 2024, the ABA released Formal Opinion 512, which explains how existing ethics duties apply when lawyers use generative AI. Those rules determine what a firm can let a chatbot do, and where a lawyer has to step back in.
What a Law Firm Chatbot Can and Can’t Do
A chatbot can collect a caller’s contact details and a short summary of what happened. It can schedule a consultation and answer basic questions about the firm.
What it can’t do is give legal advice. The line between “here’s how our firm handles this” and “here’s what you should do” is where most chatbots get into trouble.
Why the ABA Says Lawyers Still Own the Judgment Calls
Formal Opinion 512 doesn’t ban AI. It says a lawyer can’t hand off anything that needs personal judgment. That includes evaluating a claim or advising a client on next steps.
A law firm chatbot that sticks to that intake role fits neatly into that rule. Basic intake triage, such as routing a matter to the appropriate practice area, can be appropriate when the system is properly designed and supervised. The problem starts when the chatbot moves from categorizing the inquiry to independently evaluating the merits or telling the visitor what legal action to take.
What Florida’s Chatbot Opinion Requires
Florida Bar Opinion 24-1 addresses chatbots directly. It requires a clear disclaimer that the visitor is talking to an AI program, not a lawyer. The opinion also recommends screening questions that limit the chatbot’s communications when another lawyer already represents a visitor.
What Happens to What a Client Types?
The ABA also flags a real risk in Formal Opinion 512. Some generative AI systems may retain, reuse, or learn from information entered into them, while others offer different retention and privacy controls. A firm needs to understand the particular tool’s terms, privacy practices, data retention, and access rules before allowing prospective clients to enter sensitive facts. With some self-learning systems, information entered for one matter can create a risk of later disclosure or reuse in another context.
So, a chatbot that stores or reuses visitor input needs a firm to know how that data moves. If a firm can’t answer that question, the tool isn’t ready for client-facing use yet. Those questions should be settled before launch. A marketing team that also tracks legal ethics rules should be able to account for what the chatbot collects, how that information moves, and where a lawyer needs to step in.
There is also a conflict-check reason not to let a chatbot collect unlimited detail. Under Model Rule 1.18, information received from a prospective client can create confidentiality duties and, in some circumstances, conflicts even if the firm never takes the case. Intake scripts should collect what the firm needs without encouraging visitors to dump sensitive facts into the chatbot before conflicts are checked.
Why a Chatbot Should Never Replace the First Phone Call
An intake chatbot works best as a filter, not a replacement for the moment a person reaches a human. Firms that pick up the phone and return calls personally treat that first call as part of the service.
When someone spends ten minutes talking to a bot and gets nothing but a form confirmation, they don’t feel good. They need to feel like their time is respected, and a paltry “thank you for filling out this form!” does the opposite. It’s the same reason why law firm traffic often doesn’t convert in other parts of a site, just disguised as a chatbot instead of a landing page.
Fees, Billing, and the One Rule Firms Forget About AI Tools
Model Rule 1.5 covers this. A firm can’t speed up their work with AI and bill the client the amount it would have taken without it. Hourly billing needs to match the time spent, not the time it would have taken without the chatbot’s help.
This applies to intake too. The same principle applies when AI speeds up billable work after the representation begins. Efficiency does not permit an hourly firm to bill time nobody actually spent.
Setting Up a Legal Chatbot Without Creating a Compliance Problem
Start with a script that only asks for facts: who’s asking, how to reach them, and a short description of the issue. Keep every response about legal questions pointed toward booking a call, never toward an answer.
A visible way for a visitor to reach the firm directly should sit next to the chatbot, not three clicks away. Someone who wants a person, not a bot, should never have to search for that option.
Those intake decisions belong inside the firm’s larger strategy. Firms building out a full marketing plan need a clear line between automation and human contact before the chatbot goes live.
Measuring Whether a Law Firm Chatbot Helps at All
A chatbot that generates form submissions isn’t the same as one that generates signed clients. A firm that ties chatbot activity back to its intake and CRM system can see which contacts turned into real consultations.
Without that link, a firm might keep a chatbot script that looks busy on paper. But in practice, it’s driving away visitors that a real phone call with a real person would have kept.
Frequently Asked Questions About Law Firm Chatbots
Do clients need to know they’re talking to a chatbot, not a lawyer?
Yes. Florida’s opinion requires a clear disclaimer in that state, but transparency on the subject with clients is generally a good idea. A chatbot pretending to be a person creates the kind of misunderstanding bar rules are built to prevent.
Can a chatbot give a caller a rough idea of whether they have a case?
A chatbot should not independently tell a visitor whether they have a viable claim. It can collect relevant facts and route the inquiry, but an individualized assessment of legal rights or strategy should come from a lawyer.
Does a firm need a written AI policy before using a chatbot?
There is no universal rule requiring every law firm to maintain a document titled ‘AI Policy.’ ABA Formal Opinion 512 does say managerial lawyers should establish clear policies governing permissible GAI use and ensure that lawyers and staff understand the tools’ risks and limitations
What a Law Firm Chatbot Should Do
A law firm chatbot works best as a fast, well-supervised front door, not a replacement for the lawyer behind it.
The risk starts when the bot moves from “here’s how we can help” to “here’s what you should do.”
Start by writing down what the chatbot is allowed to say, then build the script around that list.
Sources
American Bar Association, Formal Opinion 512: Generative Artificial Intelligence Tools