AI Can Help You Prepare, But It Can't Replace Legal Strategy

A client sent me an email with 15 questions. Every one was a hypothetical. Every one had clearly come out of an AI chat.
I'm going to defend him, mostly.
The emails I get now look different than they did two years ago. People show up knowing what a priority date is. They know there's a backlog and roughly where they sit in it. They've already put their situation into an AI and seen the list of options.
That helps. The things I do are complicated, and there's a lot of nuance. A client who arrives with some background means we spend the strategy session on strategy instead of on how the line works.
And for the record, most of what the AI tells them is right. Usually I'm correcting small things or adding the part that depends on their facts. I use it myself. I'll put a scenario in and ask whether I've missed an angle. I'm not too proud to consider a good idea because it came from a computer.
The hard part is the 15-question emails I mentioned. A long chain of hypotheticals the client wouldn't have come up with on their own is real time, but I would not usually bill separately for time spent responding to client emails in my flat-fee structure. I also never want a client to feel like they can't ask me things. It's their life, it's their case, and they've paid me a lot to get it right.
So I answer. With that client, after a few rounds, I finally wrote: I think we've gone through enough hypotheticals. He agreed. We'd covered the options. The rest was noise.
If you're using AI to prepare for a conversation with your attorney, keep doing it. Then do one more thing. Pick the one or two questions you actually care about, and lead with those.


