@ai4lawyersTikTok
All transcripts

Ethics & professional responsibility

What to tell clients about AI when representation begins

At intake, identify a client’s prior AI use, preserve a clear record, stop unmanaged chats about the matter, and protect communications after counsel is retained.

Practice note

Privilege and work-product questions are fact-specific.

A client’s AI use before retention does not automatically establish privilege or work-product protection. Preserve the relevant facts, evaluate the governing law and the actual communications, and address any discovery or preservation obligations. After retention, do not allow client communications about the matter to migrate into unapproved tools or chats without considering confidentiality and waiver risks.

Client AI use before and after retention

FactorUnmanaged client AI useDocumented and counsel-directed process
Existing chatsThe lawyer may not know what was asked, shared, or preserved.Identify relevant use and preserve a factual record for privilege and discovery analysis.
Future useThe client continues chatting about the facts or legal theories without guidance.Set a clear boundary and explain what information should not be entered into a chatbot.
ConfidentialityPost-retention legal discussions may enter an unapproved service.Use only approved tools and protect confidential communications.

Example conversion prompt

End a client’s pre-retention AI chat about the matter

I have now retained an attorney and am ending the use of this chat for this matter.

Edited transcript

Transcript

Hey lawyers, even if you’re not using AI, your clients may be. Here’s what you can do about it.

1. Identify the Client’s Pre-Retention AI Use

First, if your client used AI before you were retained, they may be protected by the Pro Se Work Product Doctrine. If you want to know more about that, I encourage you to check out Medic Council’s page. She just did an excellent video on this topic.

2. Have the Client Log Relevant AI Use

Second, have your client log all of the use of AI involving the case, the facts, or the theories of the case. You’ll need this later to create a privilege log.

With a privilege log, the judge will know what is actually at stake on any motion to compel responses or compliance with a privilege subpoena. It is much better to say, “Your Honor, this is what’s at stake,” rather than, “Yes, my client used AI. I don’t know what’s out there.”

3. Set a Clear Boundary After Retention

Third, have your client go to each and every open chat and paste this prompt:

I have now retained an attorney and I’m ending the use of this chat for this matter.

That creates a clear line.

4. Keep Post-Retention Communications Out of Unapproved Chats

Fourth, encourage your client not to use AI. But if they insist and want to use AI, make sure they do not put any facts or theories you have discussed with them, or any confidential communications, into the chatbot. Entering those confidential communications could waive confidentiality.

Follow me for more tips and tricks on how to use AI in your legal practice.