Can AI Help Lawyers Build Stronger Arguments? Yes — Here's the Technique That Actually Works
AI is a great devil's advocate!
5/27/20264 min read
Can AI Help Lawyers Build Stronger Arguments? Yes — Here's the Technique That Actually Works
By Angeli Fitch, AI Compliance & Ethics Attorney | 20+ Years of Trial Experience | Creator of the State Bar-Approved CLE Course "AI Ethics for Attorneys"
What is the biggest mistake lawyers make when using AI?
They use it to produce work product instead of pressure-test it.
Most attorneys who have adopted AI are using it as a drafting tool — a faster way to generate a brief, summarize a document, or pull research. That's not wrong, but it misses the most powerful thing AI can do for a lawyer: argue back.
After more than two decades trying cases, I can tell you that the most dangerous moment in litigation isn't when opposing counsel attacks your argument. It's the moment before, when you've become so convinced by your own reasoning that you stop questioning it. AI can fix that — if you use it right.
What does it mean to use AI as a devil's advocate?
It means flipping the prompt. Instead of asking AI to help you build your argument, you ask it to destroy the argument you've already built.
You hand AI your motion, your legal theory, your contract position — and you instruct it to attack. To find every unsupported assumption, every weak precedent, every logical gap that opposing counsel would exploit or a skeptical judge would probe. Then you use what it finds to strengthen your position before anyone else sees it.
This is stress-testing. And it's one of the most practical, ethically sound ways a lawyer can use AI in active litigation or transactional work.
How do you actually prompt AI to challenge your legal argument?
The prompt matters. Vague instructions produce vague criticism. Specific instructions produce the kind of pointed attack that's actually useful.
Here are two prompts that work:
For a motion or brief: "You are opposing counsel who needs to win this case. Tear apart this motion for summary judgment. Identify every weakness in my legal reasoning, every case I've cited that might not hold up on closer examination, every factual gap, and every argument I haven't addressed that you would raise."
For oral argument preparation: "You are a skeptical judge who believes my client's position is weak. Ask me the hardest questions you can think of about this argument. Do not hold back."
The goal is to surface the questions you haven't answered yet — before you're standing in court being asked them.
Is this just moot court? What makes AI different?
Moot court is the right analogy, and it's actually a useful one. Law schools use moot court because forcing advocates to face hard questions before the real argument makes them better. AI does the same thing, with a few practical differences that matter for working attorneys.
It's available at 11 PM the night before a filing deadline. It doesn't have feelings to spare. It doesn't defer to you because you're the senior partner. It will identify the weakness in your argument even if you've spent three weeks building it — especially then.
For sole practitioners and small firms that don't have the luxury of a team to stress-test their work, this is genuinely significant. The playing field shifts.
What kinds of legal work does this approach apply to?
Any work where there's an opposing interest or a decision-maker to persuade. That includes:
Litigation — stress-test motions, briefs, and oral argument positions before filing or appearing. Discover what opposing counsel will argue before they argue it.
Contract negotiation — ask AI to identify every provision your counterparty's attorney is going to push back on, and why. Go into the negotiation having already thought through the other side's objections.
Client counseling — before you advise a client on a course of action, ask AI what risks you might be underweighting. A second perspective, even a synthetic one, catches blind spots.
Settlement analysis — ask AI to identify your weakest positions and your client's most significant exposure. That's the honest foundation for any settlement conversation.
Does using AI this way raise any ethical concerns for lawyers?
It can, and attorneys should be clear-eyed about that.
The confidentiality question comes up first. When you feed a brief or legal strategy into an AI tool, you're potentially sharing privileged client information with a third-party platform. Before using any AI tool for this purpose, you need to understand its data retention and training policies. Using a general-purpose consumer AI product without reviewing those terms may violate your duty of confidentiality under Rule 1.6.
The competence question comes up second. AI devil's advocate is a tool, not a substitute for judgment. If AI identifies a weakness in your argument, you still need to evaluate whether that criticism is legally sound. AI can misread precedent, miss jurisdictional nuance, or flag issues that aren't actually weaknesses in context. You are still the lawyer. The analysis is yours.
Used carefully and with appropriate vetting of the tools involved, this approach is not only ethically permissible — it reflects the kind of competent, thorough practice Rule 1.1 requires.
What's the practical takeaway for lawyers who want to try this?
Take the next argument you're preparing — a motion, a contract position, a litigation strategy — and before you finalize it, give it to AI and ask it to attack. Note what it finds. Revise. Then ask it to attack the revision.
You'll either confirm that your argument is solid, or you'll find the holes before your opponent does. Either outcome makes you a better advocate for your client.
The lawyers who are getting the most out of AI right now aren't the ones using it to write faster. They're the ones using it to think harder. That's a distinction worth understanding before you decide how AI fits into your practice.
Angeli Fitch is an AI Compliance & Ethics Attorney and trial lawyer with more than 20 years of experience. She is the creator of the California State Bar-approved CLE course "AI Ethics for Attorneys" and advises law firms and legal professionals on ethical AI adoption, compliance, and governance. She is Of Counsel at Infinity Law Group and available for speaking engagements, CLE instruction, and advisory work.