Prompt engineering for lawyers

Ask a tool like ChatGPT a question and the answer can feel like magic. Under the hood, the model does one thing over and over: it predicts the next word. Modern generative AI runs on a design called the transformer, and its key feature is self-attention, which lets the model weigh how every word in your prompt relates to every other word, much the way you do when you read.

Walk through what happens when you type "What are the consequences of copyright infringement?" First the model chops your question into tokens, small chunks of text it can work with. Those tokens pass through layers that size up what matters: one layer marks "consequences" as the key term, another ties it to "copyright infringement." Then the model writes back one token at a time, choosing each next word by probability. "The" comes first. In this context, words like "penalties," "fines," and "damages" rank high. Token by token, it builds a complete answer: fines, damages, injunctions, maybe criminal charges.

This matters for lawyers because nothing in that process checks whether an answer is true. The model produces the answer that seems most likely, and a likely answer can still be wrong. The skills on this page help close that gap. Prompting is how you change which answers the model treats as likely.

A few distinctions worth keeping straight. General chatbots (ChatGPT, Gemini, Copilot) draw on broad knowledge: versatile, but not specialized. Tools like Westlaw AI work from structured legal data and are more reliable inside their lane. And a tool is either cloud-based, meaning your prompts travel to someone else’s servers, or local, meaning they stay on hardware you control. That difference matters every time client information is involved.

These tools sometimes get legal questions wrong, and the reason is usually a mismatch: they do not think the way we think. Left alone, a model does not always understand that you want the truth, or that "I do not know" beats an invented answer. Left to their defaults, these tools favor fluent, creative answers, so accuracy takes instructions from you.

So use AI where it is strong. AI is very good at improving writing style, because style work calls for options rather than facts: it has processed more sentences than any writer alive, so let it brainstorm ten ways to recast yours. It summarizes defined text fast and well: the cases, briefs, and statutes you paste in. It generates themes, storylines, and headlines. It drafts starter text that beats a blank page. It extracts and organizes information into clean tables. And it explains and translates complex material for any audience you name.

Now the weak spots. It can misstate the law and cite cases that do not exist, in prose that sounds completely sure of itself. It handles jurisdiction-specific nuance badly. And it has no client instincts: no empathy, no judgment, no feel for what this judge, on this record, needs to hear. It is a tool, not a wand.

Generative AI works so well at first glance that it tempts you to put your pen down. Resist. You cannot use this tool, or any tool, well if you do not know what the final product should look like. Some call this the pilot era of lawyering: technology handles more of the small details while the lawyer sits at the controls, adjusting, steering, correcting. That only works if you can still fly the plane yourself.

I like a homelier picture: the bread machine. I make a lot of bread, and the machine does the heavy lifting. No kneading by hand, no waiting on the rise. But loaves still come out bad, and when they do, I know how to adjust, because I know how to make bread. If I did not, I would get the same loaf every time, good or bad, and I would be taking orders from an appliance. So decide now: do you want to be a bread machine assistant, or do you want the bread machine to assist you?

The tunes in the menu below are written with that question in mind. Each one leaves the decisions with you. The AI drafts and makes suggestions, and you decide what to use.

You owe clients competence and diligence, and you owe courts candor. Use AI to draft a paragraph and you are responsible for that paragraph as if you wrote it yourself. Competence now includes knowing how your tools work and staying reasonably current as they change. Check the rules that govern you, too: California bar guidance says lawyers should get client approval before using generative AI on a matter, and some courts require you to disclose AI use in your briefing.

Confidentiality has two exposure paths, and you need to check both. First, transmission and storage: like any cloud tool, an AI service streams your data out, holds it while working, and streams a response back, and it can be compromised at any point along that path. Second, and unique to AI: training data. Your prompts are gold to AI companies, which pay people to do all day what you do for free. Confirm your inputs will not be folded into a model’s training in a way others could extract.

Keep bias on the radar as well. These models learn from enormous datasets that include discriminatory content, and garbage in means garbage out. Read outputs with a critical eye for stereotypes and slanted framing no careful lawyer would knowingly sign.

AI output can be wrong in ways that read as right. It can miss the depth of a novel issue, mangle a doctrine, and produce a citation that does not exist, all in confident, fluent prose. So make verification a step in your process: every authority checked by a human, every quotation confirmed against the source, every legal statement traced to something you read yourself.

And make the AI help. Tell it to flag what it is unsure of. Give it standing permission to say "I do not know." Give it an exit so it never needs to invent one:

Specific, narrow instructions work better than broad, vague ones. "Draft a contract" leaves the model to guess at almost everything. Start with a concrete verb (summarize, brainstorm, rewrite, extract), deliver one point at a time, and keep the task tight. You can always go deeper on the next turn.

Then mark off your text. Delimiters separate your instructions from the text you want worked on and from your examples, so the model never confuses the document with the directions. Anything works as a delimiter if it cannot be mistaken for a word, and per the research, *** works very well.

A role tells the model how to approach the task and how to shape the answer. A research-assistant role produces neutral summary. An advocate role produces persuasion. A judge role shows you the pushback your argument will draw before a real judge supplies it. Most tasks benefit from one.

Open your prompt with a short phrase that sets the role, and add a quality adjective ("a well regarded lawyer," "a renowned legal writer") to point the model at its best material. For style work, mix favorite legal and non-legal writers until the blend sounds right.

These roles are different from Prompt Pilot’s lenses. A role here is a persona you give the AI. Prompt Pilot’s practice and role lenses describe your side of the work instead: your practice area and your role. You can use both.

These models can answer quickly by matching a pattern, or work through a problem in steps. Complex legal tasks need the second approach, and chain-of-thought prompting is how you ask for it. The easiest way to trigger it is with a few magic words. "Let’s think step by step" measurably improves results, and one wording performs even better in the research: "Let’s work this out step by step to be sure we have the right answer."

You can also list the steps yourself: first find the relevant provisions, second analyze how they apply, third conclude. Or make the model design its own plan and follow it. And when you want your work checked, do not let the AI grade first. Make it solve the problem independently, then compare:

The model will follow nearly any limits you set on the structure of its answer, so set them. Control depth with numbers: not "improve this verb" but "give me 10 versions, each using a different verb." Confine length the same way when it rambles. State facts it must treat as true: "All text within quotation marks is accurate and should not be modified." Give it a reading grade or a named audience. And name the format you want back, because the model is very good at following a pattern:

These are called few-shot prompts: you show the model a few examples of what you want, and it picks up the pattern quickly. The research guidance: provide at least four examples when you can, make them diverse, randomize their order, structure them as prompt-and-answer pairs, and include wrong answers next to right ones so the model learns the difference.

Background makes a large difference, so supply it and label it. State "The following information is context," mark it off with delimiters, and include your jurisdiction, the key cases, the bullet points from your draft. (Every tool has a context window, a ceiling on how much text it can hold at once, so spend the space on what matters.)

Then add hints. In recent studies, a short "Hint:" list of keywords significantly improved response quality. Summarizing a case? Tell the model which concepts the summary must center on:

You will rarely get the best answer on the first try, and you do not need to. The model remembers the conversation, so treat it like a junior colleague: say what it got right, what it got wrong, and what to change. Quote its own response back to it. If it seems unsure, tell it to ask you clarifying questions.

A few follow-up techniques the research supports. Phrase instructions affirmatively: say what to do, not what to avoid. Repeat key instructions at the end of a long prompt, because models weight the ending most (recency bias). Have the model generate background knowledge first, then answer on top of it. For a prompt that really matters, run five versions and compare; five is the number the research suggests. You can also combine prompt types. This one sets a role, a task, a limit, and hints in four short lines:

AI tends to produce the most typical answer: the usual structure, the usual motion, the usual sentence. That is a reasonable start but often not good enough to finish with. When you need something less typical, ask for it. "Provide the contrarian view on this topic." "Come up with three unique angles most lawyers wouldn’t use." "Brainstorm the most unusual approach for the verb in this sentence." Creativity prompts rarely produce a finished product, but they often produce ideas you would not have found on your own.

How far can this go? I was consulting with an elite appellate team on a pro bono case, impact litigation where every word counted. They had run dozens of versions of one heading and none was right. We tried a couple of generic prompts. Still flat. The case involved parody, so just for fun, with the team watching on Zoom, I typed: "Try rewriting this heading in the form of a unique poem." Silly, right? The model started producing verses, and one line carried a turn of phrase we all recognized instantly as the seed of the perfect heading. A little massaging later, we had our winner. I am not telling you to commission poetry for every brief. I am telling you to experiment, because you never know which prompt shakes the right answer loose.

Where is AI most useful in legal writing? Style, above all: "Pick better verbs in this sentence to convey [the emotion you are aiming for]." Grammar and mechanics: "Correct the punctuation in this sentence." Summarizing and extracting: "Summarize the statute and keep it to one paragraph." Starter drafts for rote sections and templates. Brainstorming: "Write a theme for two people fighting over a retirement account." Classifying and formatting: "List these cases in alphabetical order." And collaboration: run a presentation past it, ask for the weaknesses in your oral argument, or have it turn your margin notes into feedback a colleague can actually use.

The prompt below turns a sample of your writing into a style guide. After a few follow-up prompts, you will have a writing playbook for yourself or your whole team:

In the end, generative AI can be a fantastic legal writing partner. It brainstorms, helps you see a sentence differently, and organizes and extracts information. But good results come from trial and error, with you in the pilot seat: questioning, guiding, editing, and verifying. Learn what it does well, what it needs help with, and what it should not be used for.

A takeaway sentence has to do two things: include the specific details that persuade, and make sense to a reader who has not read the section below it. That is hard to do in one sentence, so the prompt asks for four options and you choose one.

The prompt uses four of the nine skills. It sets a role, states both requirements, gives an example of a takeaway you already like, separates the section with delimiters, and repeats the instruction at the end.

You start with a set of facts and have to turn them into a story. AI can help with that step. Each of the four prompts below looks at the same facts from a different angle, and none of them makes the decisions for you.

The genre prompt is meant seriously. A hard-boiled detective version will not be usable in a filing, but it can suggest a sentence rhythm or a way to open the scene that you might not find by rereading the chronology.

Rule sections are usually the densest part of a document, and that is where jargon tends to collect. The first two prompts look for it. The third is worth saving: it asks the AI to argue against you, with specific reasons, before opposing counsel does.

The chapter also states a limit. Do not use a general chatbot to find a case, a rule statement, or a quotation. Use a research tool built for that work, and verify the results either way.

Framing happens at the level of words: you present the same facts from your client’s side. A tool that can produce many versions of a sentence quickly is useful for that. The substance stays the same, and you choose the version.

Do not skip the last prompt. Ask for the three strongest arguments against you while you still have time to answer them.

Style is one of the best uses of AI that the book describes. Style work asks the AI for options, so it does not depend on the AI getting facts right. A model has processed more sentences than any writer. Even if you use none of the five versions you ask for, comparing them can teach you something about the sentence you wrote.

Each prompt below gives you options to choose from and leaves the writing to you.

Grammar and spelling checkers catch mechanical errors more consistently than a chatbot. A chatbot is useful for the explanation: why a comma belongs where it does, or why a clause is wrong, in terms you can apply the next time.

Both prompts end by asking for an explanation you can apply in your own writing.

Lawyers rarely get feedback on the design of a filing. These three prompts ask for it: what kind of chart fits your data, how to describe an image for a reader who cannot see it, and how to improve the typography.

The last technology note in the book is about improving your writing rather than producing work. Have the AI give you exercises, then check your answers against its explanation.

The second prompt is a good one to start with. Ask for a badly written paragraph and edit it. It takes about ten minutes, and no client work is at stake.