Most lawyers want to use artificial intelligence more effectively, but many hold back. The concerns are familiar: malpractice exposure, hallucinations, confidentiality breaches and the fear of relying on a tool that does not feel fully trustworthy.
In my work coaching attorneys across practice areas, I have noticed the same pattern. The lawyers who are most frustrated with AI treat it like a search engine. They type a single, broad question; receive a generic answer; and conclude that the model does not understand the problem. They either abandon AI entirely or continue using it ineffectively, which only increases their risk.
But AI is not a search bar. It is a dialogue. And the quality of the output depends entirely on the quality and specificity of the information the lawyer provides.
There is one technique I teach that consistently improves accuracy, reduces hallucinations, and helps lawyers stay firmly within their ethical boundaries. I call it the Curiosity Prompt.
The technique
It is one simple line that can be added to any prompt: “Ask me what else you need to know to give me the most accurate response.”
This shifts the lawyer from hoping the model knows what matters to directing the model to gather the relevant context.
Instead of guessing which facts or constraints to include upfront, the lawyer invites the model to ask clarifying questions. It functions very much like a supervising attorney questioning a junior associate before research or drafting begins.
The result is more precise answers based on facts, goals, constraints and nuances the lawyer may not have thought to articulate initially.
Why this reduces risk
When used correctly, this approach supports core duties of competence, diligence and supervision.
It reduces hallucinations
Hallucinations often occur when the model lacks context but produces an answer anyway. By instructing AI to ask follow-up questions before it begins, the lawyer minimizes the risk of the model guessing or filling in gaps inaccurately.
It increases lawyer control
Because the model interviews the user, the attorney remains the source of all material facts and guiding constraints. AI is not making assumptions. It is asking for direction.
It mirrors how lawyers already work
Lawyers routinely train junior attorneys to ask clarifying questions before conducting research or drafting documents. AI becomes another supervised assistant, rather than an autonomous decision-maker.
It produces more reliable starting points
AI is not meant to replace legal analysis. It accelerates early stage thinking, such as outlining issues, identifying considerations, spotting risks and summarizing unfamiliar areas of law.
A note on confidentiality
The Curiosity Prompt works best when lawyers maintain the same confidentiality standards they already apply to all technology tools.
Never enter confidential client information into a public AI system.
Instead, describe matters in generalized terms. Use the same approach you would use when discussing a case with a colleague while preserving privilege. The goal is to provide enough context for the model to ask meaningful questions while protecting sensitive details.
For example:
Not: “My client Sarah is suing her business partner for embezzling $400,000.”
But: “I am working on a partnership dispute involving allegations of financial misconduct.”
Many lawyers also adopt one or more of the following practices:
• Anonymize names and identifying details
• Change nonmaterial facts that do not affect the legal analysis
• Use a firm approved enterprise AI product with confidentiality protections
• Reserve sensitive work for tools inside the firm’s secure environment
Used this way, AI becomes a safe thought partner, while the lawyer remains in full control of the facts, the framing and the final analysis. The Curiosity Prompt does not require revealing privileged information. It simply helps lawyers ask better questions and think more strategically.
The Curiosity Prompt in practice
Here are five ways lawyers are already using this method safely and effectively.
1. Early-stage research in an unfamiliar area
Prompt: “I am taking on a probate matter for the first time. Ask me what you need to know to prepare a high-level overview of the key issues.”
The model will ask about jurisdiction, the type of proceeding, client posture, deadlines and the attorney’s level of familiarity. The result is a tailored framework, rather than a generic summary.
2. Drafting clearer client communications
Prompt: “I need to explain a complex procedural issue to a client in plain language. Ask me what else you need to know to help me draft this explanation clearly.”
AI will ask about tone, urgency, client sophistication and what the attorney wants the client to understand or do.
3. Preparing for difficult conversations
Prompt: “Here is the situation with opposing counsel. Ask me what else you need to know to help me plan my response strategy.”
Follow-up questions often reveal unspoken priorities and constraints the lawyer has not yet articulated.
4. Stress testing a legal position
Prompt: “Here is how I am thinking about positioning this argument. Ask me what else you need to know and then challenge my assumptions.”
This produces a structured list of weaknesses, blind spots and alternative approaches.
5. Drafting professional materials
Prompt: “Here is my background. Ask me what else you need to know to write a compelling professional bio.”
This often uncovers details the lawyer would not think to include but that matter for credibility and marketing.
The professional advantage
The attorneys who benefit most from AI are not the ones with the longest prompts or the most sophisticated tools. They are the ones willing to think out loud, provide context, invite clarification, iterate and supervise AI the same way they supervise human staff.
The Curiosity Prompt helps lawyers use AI the same way the profession already teaches them to work. With thoughtful questioning. With oversight. With intentional framing.
AI is not a replacement for legal judgment. It is a catalyst for clearer thinking and more efficient preparation.
When lawyers stop treating AI like a search bar and start treating it like a junior researcher who must ask clarifying questions before beginning work, they get better results with less risk.
What this means going forward
The lawyers who will thrive in the next decade are not the ones waiting for AI to feel safe. They are the ones learning how to guide it, question it, supervise it and use it as a tool that supports their expertise, rather than replacing it.
One question can change the entire experience: “Ask me what else you need to know.” It turns AI from a guessing machine into a partner in clear thinking.
Brooke Loesby is a former BigLaw attorney and the founder of Law Life Coach Inc. She helps lawyers redefine success through career strategy, leadership development and well-being initiatives.
ABAJournal.com is accepting queries for original, thoughtful, nonpromotional articles and commentary by unpaid contributors to run in the Your Voice section. Details and submission guidelines are posted at “Your Submissions, Your Voice.”
