Hacker News Reader

Comment by ss108 | Hacker News Reader

It would have taken about 1 minute to put each of them into a tool like Casetext, Lexis, WL, or Bloomberg Law to determine they didn't exist.

View on HN

To do that you first need to distrust AI, and a lot of people don't. They think of GPT like Google-but-written-in-English. That is a large part of the problem.

Well, leaving aside whether a lot of people don't distrust and whether a lawyer would or should be among those people, lawyers typically don't cite things without checking them first. I work at a biglaw firm, can't imagine it never getting caught that the cases don't exist. This sounds like a midsize firm, should be fairly similar.

There are a few small law firms/sole prop type guys, however, who I have crossed paths with for whom this kind of stupidity and carelessness would be on brand though.

Guess he was just in a rush and figured this would be one of the 2/10 times he files something without at least taking a look at the opinions first, and it ended up being a massive error.

That's not a valid excuse, though. Lawyers are paid big bucks to think, not to assume. Otherwise you could do your litigation for free by just asking interested people on Twitter. I went to law school and had to drop out due to an injury & attendant medical costs; it's a crime (as in going to jail) for me to practice law without being licensed, no matter how good my work product might be.

There are probably thousands of lawyers that thought about using ChatGPT for their profession, most of them realized it lies and never got farther than that, maybe a few hundred actually tried it out and also realized ChatGPT was lying, this is the one guy managed to swiss-cheese-model his way through.

A lot of Google's top suggested results / autogenerated answers tell you nonsense or misunderstandings as well.

Exactly. This is how AI should be used. Maybe Chatgpt hallucinates a lot, but that doesn’t mean it’s useless. It can still provide a lot of good value, you just have to filter out the crap, just like you’d do with any other tool. Especially if you’re using it in this context, you should double check its facts. I’d say it’s more the fault of the lazy lawyer that didn’t bother to verify than the fault of ChatGPT.

100% agree.

I get a little annoyed at people seeing this AI, seeing how it's not absolutely perfect, and then acting like it's horrible.

I think the expression "All models are wrong but some are useful" applies very much to ChatGPT. It's a useful tool, even if it's not perfect.

Reply on news.ycombinator.com