Best AI Prompts for Lawyers: Practical Prompts You Can Use Today
Lawyers do not need more hype about AI. They need prompts that save time, improve clarity, and reduce messy first drafts.
Used well, AI can help with research, contract review, email drafting, issue spotting, deposition prep, and case summaries. Used badly, it can waste time or create risk. The difference is often the prompt.
This guide gives you practical, professional prompts for legal work. They are written in simple language, easy to adapt, and designed for real legal workflows.
Why AI prompts matter in legal work
Legal work depends on precision. If your prompt is vague, the output will usually be vague too. A strong prompt tells the AI what role to play, what task to complete, what format to use, what limits to respect, and what questions to ask before assuming facts.
The legal industry is moving fast. In Thomson Reuters’ 2025 legal professionals report, 26% of legal professionals said they were already using generative AI, up from 14% in the prior report, and 95% expected it to become a central part of daily workflow within five years. Source
At the same time, lawyers remain cautious for good reason. The ABA’s 2024 Legal Technology Survey found that 54.4% of lawyers saw the main benefit of AI as saving time and increasing efficiency, while 74.7% identified accuracy as the biggest concern. Source
That is exactly why good prompts matter: they help you get usable work faster while keeping human review in control.
Quick infographic: AI in legal practice
FOR LAWYERS: THE BIG PICTURE
26% Legal professionals already using GenAI
95% Expect GenAI to become central within 5 years
54.4% Lawyers say the biggest benefit is time savings
74.7% Lawyers say the biggest concern is accuracy
45% Am Law 200 firms using GenAI for legal work
76% Fortune 1000 legal teams expect GenAI cost savings
The simple formula for a strong legal prompt
A useful legal prompt usually includes five parts:
- Role — Tell the AI who it is acting as.
Example: “Act as a senior litigation associate.” - Task — Say exactly what you want.
Example: “Summarize the key indemnity risks in this clause.” - Context — Give the facts, jurisdiction, audience, or document type.
Example: “This is a SaaS agreement governed by New York law.” - Output format — Ask for a clean structure.
Example: “Return a bullet list with risk level, issue, and suggested fix.” - Limits — Prevent guessing.
Example: “If information is missing, say what is missing instead of assuming.”
A strong instruction can be as simple as:
Act as a corporate lawyer. Review the clause below under Delaware law. Identify legal risk, negotiation risk, and unclear wording. Return your answer in a table. Do not invent facts. If the clause is ambiguous, explain why.
Best practices before you use AI for legal work
Before the prompt library, here are the rules that matter most:
- Do not paste confidential or privileged material into a tool that is not approved by your firm or client.
- Always verify case law, citations, dates, names, and quotations.
- Use AI for drafting, structuring, comparing, and brainstorming — not for blind legal conclusions.
- Ask the AI to flag uncertainty, missing facts, and assumptions.
- Request output in a format you can review quickly: table, issue list, checklist, or redline notes.
- Treat AI output like junior work product: useful, but never final without review.
Best professional AI prompts for lawyers
Below is a practical table of prompts you can copy, paste, and adapt.
| Use case | Best prompt |
|---|---|
| Contract review | Act as a commercial contracts lawyer. Review the clause below for legal risk, business risk, ambiguity, and negotiation leverage. Identify any unusual wording. Return the result as a 4-column table: Issue, Why it matters, Risk level, Suggested revision. Do not assume facts not stated. |
| Contract red flags | Review this agreement as counsel for the customer. List the top 10 red flags, ranked from highest to lowest risk. Focus on indemnity, limitation of liability, termination, data use, payment, auto-renewal, IP ownership, confidentiality, audit rights, and dispute resolution. |
| Clause rewrite | Rewrite this clause in plain, enforceable business language without changing its legal intent. Then provide a second version that is more favorable to my client. Briefly explain the difference between the two versions. |
| Legal research plan | Act as a litigation associate. Build a research plan for this issue: [insert issue]. Include the key legal questions, likely controlling authorities, secondary sources to consult, possible counterarguments, and facts that could change the outcome. Keep it practical and concise. |
| Case summary | Summarize this case for a supervising attorney. Use the headings: Facts, Procedural history, Issue, Holding, Reasoning, Key quote, Practical takeaway. Keep each section short and clear. |
| Compare authorities | Compare the following cases on the issue of [issue]. Show where they agree, where they conflict, and which factual differences matter most. Return the answer as a comparison table. |
| Motion outline | Draft a persuasive outline for a motion to dismiss based on the facts below. Include introduction, statement of facts, legal standard, main arguments, likely counterarguments, and a short conclusion. Use clean headings and concise points. |
| Deposition prep | Act as trial counsel. Prepare a deposition outline for [witness type]. Organize questions by topic, objective, key documents to use, admissions to seek, and follow-up questions if the witness evades. |
| Witness credibility check | Review the witness statement below. Identify internal inconsistencies, vague statements, unsupported claims, timeline gaps, and facts that should be tested in deposition. Return a bullet list with suggested follow-up questions. |
| Due diligence | Act as M&A counsel. Review the material below and produce a due diligence memo. Organize findings into Corporate, Employment, IP, Privacy, Litigation, Regulatory, and Commercial Contracts. Mark each finding as High, Medium, or Low risk. |
| Client email draft | Draft a client email explaining this legal issue in plain English. Keep the tone calm, practical, and professional. Avoid jargon. Include: what happened, what it means, risks, options, and recommended next steps. |
| Settlement analysis | Act as litigation counsel. Based on the facts below, identify the strongest liability points, damages issues, weaknesses in our case, and settlement pressure points. Then suggest a negotiation strategy with 3 possible offer ranges and the logic behind each. |
| Timeline builder | Create a case timeline from the facts below. Put events in chronological order with date, event, legal significance, and open questions. If dates are missing or inconsistent, flag them clearly. |
| Compliance checklist | Act as regulatory counsel. Turn the following regulation or policy text into a practical compliance checklist. Group items by obligation, deadline, owner, and evidence required. Keep each checklist item short and action-oriented. |
| Policy drafting | Draft an internal policy on [topic] for a [company type]. Use clear language, defined responsibilities, reporting steps, exceptions, and review frequency. Keep the tone practical rather than academic. |
| Privilege review support | Review the text below and identify statements that may raise privilege or work-product issues. Explain why each item may be sensitive and what additional facts are needed before disclosure. |
| Billing narrative cleanup | Rewrite the following time entries so they are clear, specific, professional, and client-ready. Keep them accurate. Avoid vague phrases like ‘worked on’ or ‘attention to’. |
| Document summary | Summarize this document for a busy partner in under 250 words. Then list the 5 points that matter most for negotiation or litigation strategy. |
| Issue spotting | Act as a senior attorney doing first-pass issue spotting. Based on the facts below, identify all plausible legal issues, even if some require more facts. For each issue, state what facts matter and what questions should be answered next. |
| Jurisdiction-specific analysis | Analyze the issue below under [jurisdiction] law only. If the rule differs in other jurisdictions, mention that briefly but keep the main answer focused on [jurisdiction]. Do not cite authorities you cannot verify. |
The 7 prompts every lawyer should keep ready
If you only save a few prompts, save these:
- The risk review prompt
Best for contracts, policies, and demand letters. - The issue spotting prompt
Best for early case assessment and intake. - The client explanation prompt
Best for turning complex analysis into plain English. - The comparison prompt
Best for conflicting cases, clauses, or negotiation positions. - The rewrite prompt
Best for cleaner drafting and alternate language. - The deposition prompt
Best for witness prep and cross themes. - The checklist prompt
Best for compliance, diligence, and workflow handoffs.
Prompt templates by legal workflow
1. For contract drafting and review
Use AI to speed up first-pass review, not to replace legal judgment.
Prompt:
Act as counsel for [party]. Review the contract language below. Identify legal risk, commercial risk, ambiguity, fallback language, and points to negotiate. Present the answer in a table. Then draft a revised version that better protects [party] while staying commercially reasonable.
Why it works:
- It tells the AI which side you represent.
- It asks for both analysis and revised language.
- It forces structured output.
2. For litigation and disputes
Litigators can use AI well for structure, timelines, and argument mapping.
Prompt:
Act as a litigation associate preparing for a partner meeting. Based on the facts below, identify claims, defenses, evidentiary gaps, likely motions, and the 5 most important documents we need next. Then create a short action plan for the next 14 days.
Why it works:
- It is practical.
- It focuses on action, not theory.
- It surfaces missing evidence early.
3. For legal research
AI is useful for framing research questions, but do not rely on it for unverified authority.
Prompt:
Help me think through a legal research problem. The issue is [issue]. The jurisdiction is [jurisdiction]. First, list the sub-questions I need to answer. Second, identify factual variables that could change the rule or result. Third, suggest a research order from most important to least important. Do not invent citations.
Why it works:
- It turns AI into a research planner.
- It avoids fake authority.
- It saves time before database research begins.
4. For client communication
Clients usually want clarity, not complexity.
Prompt:
Rewrite the analysis below into a client update. Use plain English. Keep it under 300 words. Explain what the issue is, why it matters, what could happen next, and what we recommend doing now. Avoid legal jargon unless necessary.
Why it works:
- It produces a usable draft fast.
- It keeps tone professional.
- It improves client readability.
5. For compliance and internal advisory work
Compliance lawyers often need clear steps, owners, and deadlines.
Prompt:
Turn the legal requirements below into a business-friendly compliance checklist. For each item, include: action required, owner, timing, evidence to retain, and risk if missed. Use short, clear language.
Why it works:
- It converts dense law into operations.
- It is useful for legal and business teams.
- It creates a format people can actually use.
Statistics lawyers should know about AI
Here are a few numbers worth knowing:
- 26% of legal professionals said they were already using generative AI in 2025, up from 14% in the 2024 Thomson Reuters report.
- 95% of legal industry professionals surveyed by Thomson Reuters expected generative AI to become central to daily workflows within five years.
- The ABA found that 30.2% of attorneys said their offices were currently using AI-based technology tools, with usage reaching 47.8% in firms with 500+ lawyers.
- In the same ABA survey, 54.4% said the biggest benefit was time savings and efficiency, while 74.7% said accuracy was the top concern.
- LexisNexis reported that 53% of Am Law 200 firms had purchased generative AI tools and 45% were already using them for legal work.
- LexisNexis also found that 76% of Fortune 1000 legal teams expected cost savings and 72% expected to complete more work in-house because of generative AI.
The message is simple: AI is already in legal practice, but lawyers still care most about accuracy, review, and safe use. Source
Common prompt mistakes lawyers should avoid
Even good lawyers can get bad AI output for simple reasons.
Mistake 1: Asking for “a summary” with no target
Bad prompt:
Summarize this.
Better prompt:
Summarize this agreement for buyer’s counsel. Focus on payment, termination, limitation of liability, indemnity, and data use. Keep it under 200 words and end with top 5 negotiation points.
Mistake 2: Not naming the audience
Bad prompt:
Draft an email about this issue.
Better prompt:
Draft a client email for a non-lawyer CEO. Use simple language. Keep the tone calm and practical.
Mistake 3: Letting the AI guess
Bad prompt:
Analyze the case.
Better prompt:
Analyze the case using only the facts provided below. If key facts are missing, list them separately and explain why they matter.
Mistake 4: Asking for final answers too early
Bad prompt:
Tell me who wins.
Better prompt:
Identify the strongest arguments for each side, the key missing facts, and the issues that will likely control the outcome.
A reusable master prompt for lawyers
If you want one flexible prompt, use this:
Act as a careful, experienced lawyer in [practice area]. I will give you a legal task. First, restate the task in one sentence. Second, identify missing facts or assumptions. Third, complete the task using only the information provided. Fourth, present the result in a clear structure with headings and bullet points. Fifth, add a short section called “Risks and uncertainties.” Do not invent facts, citations, or quotations. If something cannot be confirmed, say so clearly.
This prompt works because it forces the AI to slow down, structure the answer, and admit uncertainty.
FAQ
What are the best AI prompts for lawyers?
The best prompts are specific, practical, and structured. They clearly define the role, task, context, output format, and limits. Good legal prompts ask the AI to identify risks, flag missing facts, avoid assumptions, and present the result in a reviewable format like a table or checklist.
Can lawyers use AI for legal research?
Yes, but carefully. AI is useful for framing issues, building research plans, comparing arguments, and simplifying dense material. It should not be trusted for unverified case citations or final legal conclusions without independent review.
What is the best prompt for contract review?
A strong contract review prompt is:
“Act as counsel for [party]. Review this clause for legal risk, business risk, ambiguity, and negotiation leverage. Return the answer in a table with Issue, Why it matters, Risk level, and Suggested revision. Do not assume missing facts.”
Can AI write legal documents?
AI can help draft first versions of contracts, letters, policies, checklists, and internal summaries. But the final document must still be reviewed by a lawyer for accuracy, jurisdiction, strategy, and ethics.
What are the risks of AI for lawyers?
The biggest risks are inaccurate output, fake citations, confidentiality problems, overreliance, and poor factual assumptions. The ABA found that accuracy was the top concern among surveyed lawyers.
Should lawyers put confidential client data into AI tools?
Only if the tool is approved for that use under your firm’s policies, client obligations, and applicable privacy and confidentiality rules. If not, use anonymized facts or avoid uploading the material entirely.
Are prompts better than general instructions?
Yes. A prompt is simply a better instruction. The more clearly you define the legal task, the more useful the result will be.
Final takeaway
AI is most useful for lawyers when it helps produce a better first draft, a faster issue list, a cleaner summary, or a sharper set of questions. It is not a substitute for legal judgment. It is a drafting and thinking tool.
The best legal prompts are not fancy. They are clear.
If you tell the AI:
- who it is,
- what you need,
- what law or context matters,
- what format to use,
- and what it must not assume,
you will usually get better legal work product.
Sources
