In the legal environment, the stakes are higher when prompting artificial intelligence to provide quick definitions, brainstorm ideas, or generate polite email responses.
AI can, and does, make mistakes, so you need to craft each instruction carefully. When used thoughtfully, AI legal prompts can streamline repetitive tasks, speed up research, and help your team produce more consistent, accurate work — whether that’s internal case summaries, initial contract drafts, external client communications, etc..
AI tools for law firms go beyond simple automation. They serve as active collaborators in drafting contracts, analyzing case law, and even predicting litigation outcomes. But getting the most from these tools requires a disciplined approach to prompt design, model selection, and data security.
This article will show you how to choose the right AI tools for lawyers, construct prompts, and offer you ready-to-use prompts.
What are the best use cases of ChatGPT prompts?
ChatGPT has the potential to transform legal work. From research and analysis to streamlining administrative tasks, AI can make your firm work smarter and more efficiently. However, the technology also has limitations, so you’ll need to know where it excels to get the most out of it.
Research and analysis
ChatGPT’s ability to analyze a large volume of documents can save you time during research. Especially during the early phases of a case, ChatGPT is effective at helping identify relevant precedents, spot patterns across documents, and summarize cases.
The technology is particularly good at finding relevant statutes, case law, and regulations. This ability makes it helpful with regulatory reviews, preliminary case assessments, and jurisdictional comparisons. However, an attorney should still conduct a detailed analysis.
Drafting
Drafting documents can be a time-intensive task, making it an ideal candidate for AI assistance. You can use ChatGPT to generate first drafts of contracts, engagement letters, pleadings, and client communications.
It works best when you give it clear parameters, such as an established structure, standardized clauses, and similarities across multiple documents. However, you’ll always want to review AI-generated documents, especially ones requiring customization.
Summarization
ChatGPT can provide quick summaries of lengthy documents, such as deposition transcripts, discovery documents, case files, and medical records. Reviewing these materials manually usually takes hours. With AI, you can summarize them in minutes.
You can also ask ChatGPT to extract key information from documents you need. For example, you could ask it to identify relevant dates or organize events from many materials into a single timeline.
Again, you’ll need to review the results for accuracy. But the initial summaries will save your paralegals and associates a lot of time compared to traditional workflows.
Administrative tasks
Administrative overhead can consume a lot of your staff’s time and resources. ChatGPT can streamline these tasks by, for example, preparing meeting agendas, formatting citations, drafting case status updates, and creating checklists.
You can also use AI to suggest more efficient workflows. As a result, your attorneys and staff will have more time to focus on high-value tasks that bring in additional revenue to your firm.
Limitations of ChatGPT for lawyers.
You need to recognize that ChatGPT has some limitations within a legal context. AI generates text based on patterns, so it shouldn’t be relied upon to give legal advice. An attorney has to review whatever materials it produces, especially to catch “hallucinations,” which are instances of plausible-sounding but factually incorrect information.
There are also technical limitations to what ChatGPT can do. For example, it doesn’t have built-in access to current case law databases. You’ll also need additional tools to verify citations and holdings.
Finally, pay attention to confidentiality concerns. You should never input privileged client information or case-specific data into the public ChatGPT interface. Doing so may risk violating attorney-client privilege and professional conduct rules.
Create clear protocols for your firm about information that can and cannot be shared with AI and ensure all attorneys and staff understand them.
How to decide which task to use ChatGPT for
ChatGPT works best when you assign it tasks that fit its capabilities. When deciding what to use it for, focus on the following factors:
- Repetitive tasks: AI is ideal for routine tasks, like creating case status templates or meeting agendas. Attorneys should focus on anything that requires creative problem-solving.
- General legal work: Tasks that only require a broad understanding of legal principles or document structures fit well with what ChatGPT can do. Attorneys will still need to provide factual analysis for specific cases.
- Internal-facing tasks: While ChatGPT can work for some client-facing tasks, like routine updates and newsletters, it’s best leveraged for internal tasks, such as draft outlines, research summaries, and administrative documents.
- Initial drafts: AI is good at generating initial drafts of documents, especially ones that work to a template. You’ll want an attorney to review the output and provide additional context or information.‍
- Non-confidential information: Only use AI tools like ChatGPT for tasks that use publicly available information. Inputting confidential data risks violating professional codes of conduct.
15 ChatGPT prompts for day-to-day legal tasksÂ
The following prompts will show you how law firms can use AI to boost efficiency and client satisfaction. Each one integrates best practices, balances detail with confidentiality, and encourages structured output formats to ensure clarity and usability.
1. Summarizing Case Law
Example: "You are a legal research assistant. Summarize the attached appellate decision, focusing on key facts, legal issues, holding, and reasoning. Use bullet points and plain English suitable for client-facing communication."
AI can assist attorneys in quickly distilling complex judicial opinions into actionable summaries for client memos or internal analysis. This type of prompt can save you time and provide your clients with the necessary information in plain language.
Tip: Avoid sharing full case dockets directly. Instead, extract and redact essential portions before input. Always request output in a clear format (e.g., bullet points) to simplify review and client discussion.
2. Drafting Contracts
Example: "Draft a standard settlement agreement and release for a personal injury claim. Include typical confidentiality clauses and standard release terms, but flag any provisions that might require jurisdiction-specific adjustments."
AI can quickly generate initial drafts for common legal documents, such as settlement agreements and releases, client intake and authorization forms, and retainer agreements. This gives attorneys a valuable head start and highlights sections that require detailed legal scrutiny.
Tip:
Provide key details about the type of agreement or form and standard clauses you need, but avoid including sensitive client information or specific settlement terms in your prompts. Always thoroughly review and customize AI-generated drafts before finalizing documents for client use.
Dig Deeper: Explore how AI can also be used for legal document summarization
3. Legal Memo Preparation
Example: "Prepare a concise legal memo summarizing available insurance policies and coverage limits relevant to a recent auto accident claim involving multiple defendants."
Alternatively:
"Draft an internal Demand Package Memo outlining the key arguments, injuries, and settlement value for use in preparing a personal injury settlement demand letter."
These prompts help attorneys quickly compile structured overviews, providing a valuable starting point to expedite internal decision-making and strategic case planning.
Tip: Clearly define the topic scope and jurisdictions. Always verify citations and supplement AI outputs with up-to-date primary sources.
4. Deposition Preparation
Example: "Generate 20 deposition questions for a witness in a personal injury auto accident case, focusing on facts related to liability, medical damages, and financial losses."
AI can help attorneys build comprehensive question sets that don't overlook any key topics during depositions, saving preparation time while improving thoroughness.
Tip: Avoid including personal client details. Use generalized facts to set the scenario while safeguarding sensitive information.
5. Statutory Comparison
Example: "Compare the Fair Labor Standards Act with California's wage and hour laws. Highlight key differences in overtime pay, exemptions, and penalties."
AI can expedite statutory comparisons across jurisdictions, enabling attorneys to quickly identify key legal distinctions for advisory memos or compliance reviews.
Tip: Specify which statutes and jurisdictions to compare. Cross-check AI summaries against the actual statutes before relying on the output.
6. Client Communication Draft
Example: "Draft a client email explaining next steps after a denied Social Security disability claim. Use plain English, suitable for a reading level of 8th grade or below."
AI can help simplify complex legal concepts into language that clients can easily understand, thereby improving communication and fostering client trust.
Tip: Keep input generalized to avoid sharing client identifiers. Review tone and content for legal accuracy before sending to clients.
7. Citation Formatting
Example: "Verify the following Bluebook citations for accuracy based on the 21st edition. Correct any formatting or reference errors."
AI can speed up tedious citation reviews, ensuring proper formatting and consistency across legal documents while freeing up attorney time for more substantive work.
Tip: Input citations as plain text for optimal parsing. Always conduct a final manual check to confirm correctness.
8. Medical Record Summarization & Case Outcome Prediction
Example:
"Summarize key medical records for a client involved in a car accident, highlighting injury severity, treatment history, and prognosis."
Alternatively:
"Analyze settlement data from previous slip-and-fall cases involving similar injuries to predict likely outcomes and settlement ranges for a current claim."
AI helps attorneys quickly distill lengthy medical documents into concise, actionable summaries and use historical data to anticipate case outcomes, thereby improving negotiation strategies and case planning.
Tip:
Specify injury types or the particular data set to guide the AI accurately, but avoid inputting sensitive personal information or confidential settlement amounts. Always verify accuracy and relevance before relying on AI-generated insights in legal decision-making.
9. Legal research query formulation
Example: “I’m researching if a landlord can be held liable if a tenant’s dog bites someone in a building’s common area in Florida. Provide a list of Boolean queries and search terms, including case law topics and key numbers, optimized for LexisNexis.”
ChatGPT is good for searching legal research databases. It can suggest Boolean operators, alternative terminology, relevant legal doctrines, and key search terms, while taking into account jurisdiction-specific considerations.
Utilizing ChatGPT for legal research query formulation is especially useful if you’re researching legal areas you’re less familiar with.
Tip: Begin with a broad prompt to establish a baseline. Then, refine your query based on the initial results. If you get too few results at first, you can ask ChatGPT to suggest variations on your prompt. If you get too many, specify specific jurisdictions and legal theories to narrow the focus.
10. Document review checklists
Example: “Create a due diligence checklist for commercial real estate purchase agreements. Include title review, financing contingencies, and closing conditions. Organize the checklist chronologically by review stage and include specific tasks for verifying each section.”
ChatGPT can be used to create detailed checklists for different document types, such as loan documentation, employment agreements, intellectual property assignments, or regulatory compliance.
Customizable review checklists not only ensure your attorneys don’t overlook key details, but they’re also useful for due diligence, contract reviews, and compliance audits.
Tip: Whenever you produce a checklist you’re happy with, save it as a template. You can then use that template for future checklists and ask ChatGPT to modify it for different practice areas or to reflect legal changes.
11. Timeline and chronology creation
Example: “Create a timeline for a breach of contract case using the attached documents. Put dates in the left column, events in the middle column, and source documents in the right. Highlight critical deadlines in bold.”
ChatGPT can organize essential facts from a multitude of documents into a coherent and easy-to-understand timeline. For example, you can use AI to analyze pleadings, discovery responses, communications, and depositions and arrange the key dates into a narrative structure.
For complex cases with lots of records and multiple parties, creating a simple timeline is especially helpful.
Tip: Use ChatGPT to create different versions of the same timeline for different uses and audiences. For example, create simplified ones for clients, detailed timelines for attorneys, and graphical versions for trial exhibits.
12. Pleading and motion drafting
Example: “Create a motion to dismiss because of a lack of jurisdiction. The defendant is a corporation headquartered in Delaware, being sued for breach of contract in a California state court. The corporation has no physical presence in California, and the contract includes a choice of law provision specifying Delaware.”
ChatGPT can be a great tool for generating initial drafts of motions and pleadings based on standard legal frameworks. You’ll still want an attorney to review these drafts to make sure they’re accurate and jurisdiction-compliant. But they can save a substantial amount of time during the initial drafting phase.
Tip: To get the most accurate results, specify the jurisdiction, relevant facts, and applicable legal standard to the motion in your prompt. Never file a motion generated by ChatGPT without attorney review first, especially since it may produce errors with jurisdiction-specific rules.
13. Fee agreement and engagement letter drafting
Example: “Create a fee structure for a personal injury case in Georgia. The contingency fee should be 33% prior to litigation and 40% if litigation is filed. Include provisions for expenses, termination rights, scope of representation, client responsibilities, and any others required by the State Bar of Georgia.”
ChatGPT is useful for generating client intake documents like engagement letters and fee structures. You can draft fee agreements that incorporate different structures (i.e., hourly, contingency, flat fee), practice areas, and jurisdictions.
Plus, you can ask ChatGPT to incorporate ethical and state bar-required provisions for your practice area.
Tip: While ChatGPT is getting better at understanding state bar requirements, you should always have an attorney review the document for accuracy. Also, input your firm’s policies on costs, billing, and client communication, so ChatGPT can customize intake documents that are unique to your firm.
14. Opposing counsel argument anticipation
Example: “Take on the role of opposing counsel in a case where my client is being sued for negligence in a slip-and-fall case. The plaintiff is claiming serious hip injuries, but surveillance images show them participating in physical activities inconsistent with their claimed injuries. What arguments could overcome this evidence? Explain potential weaknesses in my own defense.”
Anticipating opposing counsel’s arguments is key to effective representation. ChatGPT can help by roleplaying as opposing counsel, so you can stress-test your own arguments and identify potential gaps in your or the opposing side’s evidence.
The result is that you’ll be better prepared for trial or mediation and won’t be caught off guard.
Tip: You should provide as much context as possible, but avoid inputting confidential information, which could present ethical issues. While ChatGPT can assist with your preparation, it shouldn’t replace legal analysis.
15. Court filing requirements summary
Example: “Draft a summary of the filing requirements for a civil complaint in the United States District Court for the Central District of California. Include required document formats, filing fees, service requirements, deadlines, electronic filing procedures, and any district-specific rules that differ from the Federal Rules of Civil Procedure.”
ChatGPT can be used to summarize court filing requirements, such as checklists that include local rules, procedural requirements, and other practical considerations. If you’re handling a case in a new jurisdiction, then these district-specific summaries can be especially useful.
Tip: Court procedures are frequently updated, and there’s a risk that ChatGPT may rely on outdated information. Verify with local counsel or via the court’s official website that anything AI-generated is consistent with the most current court procedures, local rules, and standing orders.
How to Construct an Effective ChatGPT Prompt
Effective prompting is both an art and a science. The most reliable formula involves three key components:
- Intent: Clearly state what you want the AI to achieve (e.g., "summarize this deposition transcript").
- Context: Provide necessary background without disclosing privileged information (e.g., "This is a family law custody case involving interstate jurisdictional questions").
- Instruction: Specify format, tone, or constraints (e.g., "Summarize in bullet points using plain language suitable for a client update").
Avoiding Common Prompting Pitfalls
Assigning the AI a persona can help it better understand your expectations. For instance: "You are a senior associate attorney specializing in employment law." This guides the model to adopt appropriate language, depth, and legal accuracy.
One significant risk here is overloading the AI with too much or irrelevant information. Keep prompts clear, concise, and structured. Avoid feeding privileged or confidential details directly into ChatGPT prompt fields, especially in unsecured environments.
Another pitfall is recency bias. LLMs may overemphasize recent events or data points. Mitigate this by anchoring prompts with specific instructions, such as "Limit your analysis to case law prior to 2023 unless otherwise noted."
This structured approach reflects how AI is being used in the legal space: not as an autonomous legal expert, but as a competent assistant when guided correctly.
Understanding Ethics and Confidentiality in Attorney AI Prompts
You should never overlook ethical considerations when working with AI, especially in the legal field. Attorneys are bound by rules of professional conduct, including strict confidentiality and the duty to safeguard client information.
Law firms must be vigilant about how AI systems handle sensitive data. This includes:
- Avoiding direct input of privileged information into public AI models.
- Using anonymized or redacted data sets whenever possible.
- Leveraging private, secured AI environments with end-to-end encryption.
- Reviewing AI-generated output before client delivery to ensure accuracy and compliance.
Firms should also stay updated on applicable regulations, such as the ABA's recent guidance on the ethical use of AI, as well as jurisdiction-specific privacy laws like GDPR or HIPAA, if applicable.
Dig Deeper: What level of legal advice can AI provide?
How to Construct an Effective ChatGPT Prompt
Effective prompting is both an art and a science. The most reliable formula involves three key components:
- Intent: Clearly state what you want the AI to achieve (e.g., "summarize this deposition transcript").
- Context: Provide necessary background without disclosing privileged information (e.g., "This is a family law custody case involving interstate jurisdictional questions").
- Instruction: Specify format, tone, or constraints (e.g., "Summarize in bullet points using plain language suitable for a client update").
Avoiding Common Prompting Pitfalls
Assigning the AI a persona can help it better understand your expectations. For instance: "You are a senior associate attorney specializing in employment law." This guides the model to adopt appropriate language, depth, and legal accuracy.
One significant risk here is overloading the AI with too much or irrelevant information. Keep prompts clear, concise, and structured. Avoid feeding privileged or confidential details directly into ChatGPT prompt fields, especially in unsecured environments.
Another pitfall is recency bias. LLMs may overemphasize recent events or data points. Mitigate this by anchoring prompts with specific instructions, such as "Limit your analysis to case law prior to 2023 unless otherwise noted."
This structured approach reflects how AI is being used in the legal space: not as an autonomous legal expert, but as a competent assistant when guided correctly.
Key Points to Remember
Creating effective AI legal prompts requires a disciplined framework:
- Select LLMs that offer the best legal AI tools for your firm's needs.
- Use the intent + context + instruction formula for structured prompts.
- Assign appropriate personas, and define the desired output format.
- Avoid disclosing sensitive or privileged data in unsecured AI systems.
- Remember your ethical and confidentiality obligations.
- Continuously review and supervise AI-generated content before client use.
Although AI improves efficiency in the legal field, it can't replace the crucial ethical considerations and judgment calls of human lawyers. Thoughtful prompt design allows law firms to improve accuracy, reduce manual workload, and deliver more consistent client satisfaction.
