How to Ethically Use AI Under the Florida Bar’s Professional Conduct Rules

Spoiler: You still have to be a lawyer about it.

So... Can I Use AI to Write My Briefs?

You’re an overworked Florida attorney. You’ve got 10 motions due, a deposition in two hours, and that one client who calls just to hear themselves talk. Then someone whispers the magic letters: AI.

You hear rumors:“GPT can draft your motions!”“AI does case law research in seconds!”“I heard someone used it to write a whole appellate brief! 👀”And somewhere in your caffeine-addled brain, a question forms: "Is this... ethical?"

Welcome to the ultimate clash of hype vs. the Florida Bar. Buckle up.

🚨 Wait, Can Lawyers Even Use AI?

Yes. But just like with paralegals, interns, or that one associate who insists on writing everything in passive voice, you’re responsible for what it does in your name.

Let’s take a tour of how AI use intersects with Florida’s Rules of Professional Conduct, minus the fearmongering.

📜 Rule 4-1.1 – Competence

Translation: Don’t be lazy. Or reckless. Or both.

The rule says you have to provide competent representation. This includes keeping up with “the benefits and risks associated with relevant technology.”

So if you're using AI to research, write, or analyze, you’d better understand:

• What the tool is doing

• What its limits are

• Whether it’s making stuff up (spoiler: it might)

• 
  

🚫 Do Not: Paste ChatGPT’s response into your filing without reading it. Ever.✅ Do: Treat it like a law clerk. One that might hallucinate case law while sounding extremely confident.

🔒 Rule 4-1.6 – Confidentiality of Information

AKA: Don’t feed your client’s secrets into the algorithm.

Using an AI tool that sends data back to a server? You just told the cloud about your client’s trust dispute and their weird habit of forwarding 3 a.m. emails about “justice.”

The rule: You must not disclose confidential info unless your client consents or an exception applies.

Tip: If you're using a public AI tool (like ChatGPT, Claude, etc.), do not input confidential info unless:

1. You’re sure it doesn’t get stored or used for training

2. You’ve gotten client consent

3. You enjoy explaining things to the Bar

   
   

👀 Rule 4-5.1 – Responsibilities of Partners and Supervisory Lawyers

AKA: You break it, you bought it.

If you’re a partner or supervisor, and your junior associate—or your AI—blows a deadline, files a fake case, or cites “Brown v. Boardwalk Empire (11th Cir. 2023),” guess who’s on the ethics hook? You are.

Even if AI drafted it. Even if someone else hit submit. Even if you were “super busy that day.”

🧠 Rule 4-5.3 – Nonlawyer Assistants

AI isn’t a lawyer. But you already knew that.

This rule governs your relationship with nonlawyers you employ, retain, or associate with.

You know—like your paralegal, your contract brief writer, or the machine that says “as an AI language model, I cannot provide legal advice…”

AI tools fall under this rule, meaning:

• You’re required to supervise their work

• You must ensure their output complies with your ethical duties

• You’re responsible for their mistakes

  
  

Yes, that includes bad citations. Yes, that includes confidential info. Yes, even if you really thought it was reliable this time.

🎯 Rule 4-3.3 – Candor Toward the Tribunal

Let’s talk about fake cases.

In case you missed it, several lawyers across the U.S. have already been sanctioned for submitting AI-drafted filings full of fabricated citations.

Florida has not (yet) dropped the ethics hammer here, but Rule 4-3.3 makes this clear:

• You must not knowingly make false statements to the court

• You must not cite cases that do not exist

• If you discover you did, you must correct it

  
  

So, if your AI cites “The Rock v. Netflix (Fla. 5th DCA 2018)” and you can’t find it? Neither can anyone else. That’s not clever, it's an ethics time bomb.

🧾 Rule 4-1.5 – Fees and Costs

Don’t bill your AI time like it’s your time.

If you charge clients for AI-generated work, be transparent. You can’t bill for an hour of “legal analysis” if all you did was copy-paste a GPT draft and add an Oxford comma.

Also: you can’t just slap AI fees onto the invoice like “mysterious black-box sorcery — $250/hr.” Your billing must be reasonable and clearly explained.

✅ When AI Can Help (Without Tanking Your Bar Standing)

Let’s be real: AI can be an incredible tool for streamlining your practice.

Here’s when it makes sense to use it:

🚀 Use AI For...	🧨 But Don’t Use It For...
Drafting outlines	Filing without reviewing
Brainstorming legal arguments	Submitting unverified case law
Summarizing discovery docs	Analyzing confidential client info
Creating blog content (like this 🤫)	Avoiding your own professional judgment
Research prompts (with verification)	Replacing lawyers on the cheap

🛡️ AI Use in Real Life: An Appellate Hypo

You’ve just won at trial. Opposing counsel files a dense 50-page initial brief with 3 string cites per paragraph. You turn to AI for help summarizing the issues. Great!

But now it starts suggesting arguments that weren’t preserved, invents “binding” precedent from the 12th DCA (uhhh...there is no 12th DCA in Florida btw), and footnotes every sentence. Cool it.

Use it to get ideas. Use it to check blind spots. But never use it as your lawyer brain.

⚖️ Bottom Line: You’re the Lawyer. AI Is Just a Tool.

Florida’s ethics rules are tech-agnostic. They don’t ban AI. They just expect you to know how to use it without abdicating your professional responsibilities.

So don’t panic. Don’t overhype. And don’t delegate your duty of candor to a chatbot.

🧠 Final Takeaway:

Use AI like you’d use a clever—but unreliable—intern:

Helpful in bursts. Dangerous without supervision. Never the one signing the pleading.

📚 This article is for educational purposes only. It is not legal advice, and no AI tools were harmed in its creation. (Okay, maybe slightly overworked.)