AI agents for law firms: 2026 guide
AI agents handle intake, scheduling, drafting, and follow-up for law firms. See what they do, what they cost, and which fit a small practice.
AI agents run legal work for a law firm: intake, scheduling, document drafting, contract review, and client follow-up. They free lawyers from repetitive admin so billable work gets done faster. Small and solo firms adopt them to compete with larger practices. Cost is modest, and off-the-shelf tools cover most needs. The field is young, so most of what you read is hype. This is the plain version.
TL;DR
- An AI agent completes a task end to end. Legal research software only answers a query.
- Firms use agents for intake, scheduling, drafting, review, and follow-up, and that is where the billable time comes back.
- Solo firms are the laggards and the biggest opportunity, because adoption rises with firm size.
- Confidentiality and ethics are real, and the bar has now issued formal guidance on them.
- Cost is modest for off-the-shelf tools. The expensive suites are built for large firms, not solo practices.
Which AI agents are best for a law firm?
The truthful answer: no single one, and picking is done per job, not per brand. Legal-specific suites (Westlaw, Lexis+ AI, Harvey) are tuned for research and drafting. General agents (OpenAI, Anthropic Claude, Copilot) handle the admin a lawyer should not touch. Small firms rarely need the big legal suites, because the seat cost assumes a large practice. Off-the-shelf general agents cover intake, scheduling, and follow-up for a small fraction of the price.
| Tool type | Best at | Typical cost | Fits a small firm |
|---|---|---|---|
| Legal suites (Westlaw, Lexis+ AI, Harvey) | Research and domain drafting | $225 to $1,200 per user per month | Rarely, priced for large practices |
| General agents (Claude, ChatGPT, Copilot) | Intake, scheduling, follow-up | Modest monthly seat plus usage | Yes, covers most repetitive work |
The line that matters for an owner: a chatbot saves you typing, an agent saves you the task.
What are AI agents for law firms?
An AI agent is software that completes a task without you steering each step. It reads the input, reasons about it, takes an action, and reports back. For a law firm that means a new intake form that qualifies the matter and books the consult, a draft of a standard contract clause, or a follow-up reminder that never slips. It is distinct from AI legal research software like Westlaw or Lexis+ AI, which search and summarize but do not run the workflow.
The difference matters. For the broader picture of what these tools do beyond a law practice, see AI agents for small business. A general agent fits any service business. A legal tool adds the domain layer on top.
How do AI agents help a law firm day to day?
Agents take the rote work off the attorney: intake, scheduling, conflict checking, document drafting, contract review, and client follow-up. The hours that used to vanish into admin come back to billable work. Adoption reflects that promise. The American Bar Association survey found 30.2 percent of attorneys now use AI-based tools, up from 11 percent a year earlier. Yet the gain splits unevenly by firm size: 18 percent of solo attorneys use AI, while 46 percent of firms with 100 or more attorneys do. The solo practitioner is where the upside sits.
I run an agency, and the automation that changed my week was not the flashy one. It was putting follow-up on autopilot so no warm lead went cold. My own outreach system pushes roughly 200 form submissions a day through automated follow-up, and the same pattern transfers to a firm’s intake and client care. Automate the one repeating task that quietly costs you, first.
A practical spread of the day-to-day work looks like this:
| Task | What the agent does | Who it frees |
|---|---|---|
| Client intake | Qualifies the matter, captures facts, books the consult | Paralegal and intake staff |
| Scheduling | Books, reschedules, sends confirmations | Front office |
| Document drafting | Produces first drafts of standard clauses and letters | Attorney |
| Contract review | Flags risky clauses for a lawyer to check | Attorney |
| Client follow-up | Keeps the thread warm, chases missing documents | Everyone |
| Conflict checking | Screens new matters against existing ones | Attorney |
How much do AI agents cost a law firm?
The price depends on which tier fits the firm. Off-the-shelf general agents run on a modest monthly seat with usage-based topping. The industry estimates for the big legal suites are far higher, roughly $1,000 to $1,200 per lawyer per month for Harvey, with seat minimums aimed at large firms. CoCounsel lands near $225 to $400 per user per month. The real cost is not the sticker. It is the setup and the upkeep when an app or a rule changes, and the training so the agent knows the firm’s own documents.
| Tier | Example | Estimated price | Who it fits |
|---|---|---|---|
| General agent | Claude, ChatGPT | Modest seat plus usage | Solo and small firms |
| Mid legal suite | CoCounsel | $225 to $400 per user per month | Growing firms |
| Enterprise legal suite | Harvey | $1,000 to $1,200 per lawyer per month | Large firms |
Judge the price against the task you stop doing, not against the tool next to it.
Can off-the-shelf AI agents cover a law firm, or is custom required?
Off-the-shelf general agents cover most of the repetitive work a small or solo firm does: intake, scheduling, drafting scaffolds, and follow-up. Custom work adds value only where the firm has a distinctive process or needs tight control over outputs, and even then it usually starts from the same off-the-shelf core. Nothing in this post is a pitch to overpay for bespoke tooling. Start with the general agent, point it at one repeating task, and go custom only when a workflow proves worth automating further. Newness is not a weakness to hide. Anyone selling a decade of agent experience inside law firms is selling you something.
I would run the first month on a stock agent before I ever paid for custom work. If you want it built and kept running instead of owned, the AI services page covers the managed route. If you prefer to run it yourself with the prompt patterns already mapped, the skills bundle is the starting point.
Ethics shape the whole picture, and the ABA ethics guidance on generative AI sets the floor: confidentiality applies to an agent the same as to a clerk, and a lawyer stays responsible for what the tool sends out. Adoption without that knowledge is how a firm overreaches.
Do law firms need to worry about confidentiality when using AI agents?
Yes, and the bar has now said so in writing. Legal ethics require confidentiality under Model Rule 1.6, and the duty applies to AI the same way it applies to a clerk. The ABA ethics guidance, Formal Opinion 512, holds that a lawyer must protect client information when using generative AI, including picking tools that keep data secure and telling a client where the risk sits. The practical version for a small firm: use a tool with a clear data policy, avoid pasting privileged documents into a public chat, and keep a human accountable for what the agent sends out. The human stays responsible for the judgment no matter what the tool does.
Are AI agents allowed under legal ethics rules?
They are allowed, with conditions. No bar has banned the use of AI in practice. The rules impose duties instead: competence, confidentiality, and supervision of whoever does the work, human or machine. The ABA ethics guidance, Formal Opinion 512, frames the lawyer’s duty to understand the tool’s limits, keep client information confidential, and review the output before it leaves the firm. An agent that drafts is acceptable. An agent that answers a client unsupervised, without review, is the edge that gets lawyers in trouble.
What is the difference between an AI agent and AI legal research software?
Legal research software finds and summarizes law. An AI agent completes a task. Westlaw, Lexis+ AI, and Harvey answer a research question or draft a first cut. An agent takes an intake form, checks conflicts, books the consult, and sends the confirmation without you in the loop. One is a better search. The other is a worker. Confusing the two is how firms overpay for a research tool and never automate the work that actually eats the week.
If you want help finding that first workflow, a discovery call maps it in about thirty minutes.
Related: AI agents for small business
Related: best claude code skills
Related: how to create claude code skills
FAQ
Is Claude or ChatGPT better for lawyers? Neither wins outright, and the right pick is per job. Both are strong general agents for the admin a lawyer should offload, intake, scheduling, and follow-up. Try one on a single repeating task, keep privileged documents out of any public chat, and judge it on that job before you standardize.
Is there a ChatGPT for legal? Yes, the legal-specific suites are the closest thing: CoCounsel, Harvey, and Lexis+ AI are tuned for legal research and drafting. General ChatGPT still handles the firm’s admin well, so many small firms pair a general agent for operations with a legal tool only when research volume justifies the seat cost.
Do law firms need to worry about confidentiality when using AI agents? Yes. Model Rule 1.6 confidentiality applies to an AI agent the same as to a clerk, and ABA Formal Opinion 512 says a lawyer must protect client information when using generative AI. Use a tool with a clear data policy, keep privileged documents out of public chats, and keep a human accountable for whatever the agent sends.
Are AI agents allowed under legal ethics rules? Yes, with conditions. No bar has banned AI in practice. The rules impose duties of competence, confidentiality, and supervision. An agent that drafts under review is fine; an agent that answers a client unsupervised is the edge that gets lawyers in trouble.