Live data from Hacker News

Claude for Legal

github.com

141–150 of 204 posts

Re: Claude for Legal

#141

I wonder why there isn't much for real estate in this package. Are they perhaps working on a totally different real estate project. I am in that space and very nervous about it getting wiped out by anthropic or openAI. To be honest, I am not sure why they still haven't make a big play for that industry.

It would be silly to assume they're not making a real estate product, lol.

Re: Claude for Legal

#142
post #3

It seems like they ripped out Lexis, which is probably one of the most important tools for lawyers: https://github.com/anthropics/claude-for-legal/pull/5 .

Can't we just use the version of the codebase from before that change?

Yes, they are just text-based "skills" that you are free to change to your liking.

Re: Claude for Legal

#143
post #7

As a lawyer, I'm excited about this, but there are two roadblocks that I'm not sure how Anthropic will navigate: (1) For non-lawyers who use these skills/connectors/whatchamacallits to try to get legal advice, their communications are not protected by attorney-client privilege. This will absolutely bite some people in the ass. (2) If a lawyer uses this with confidential client information (which, to the uninitiated,…

Wonder if people who decide to represent themselves would be allowed to use this service “live” on the courtroom. Usually people who represent themselves fall flat because they don’t know when it’s appropriate to invoke what and at time appear to putting legal words together that don’t make sense in a given context. This would certainly help these people, if allowed, and they don’t go off the rails.

Re: Claude for Legal

#144

I wonder why there isn't much for real estate in this package. Are they perhaps working on a totally different real estate project. I am in that space and very nervous about it getting wiped out by anthropic or openAI. To be honest, I am not sure why they still haven't make a big play for that industry.

The biggest issue in the real estate world is dealing with local state, county, and municipality governments and each one's unique ordinances, etc. While I think LLMs will be great at pulling data from a variety of disparate sources (ie, records databases), or even automating interacting with them, they are not going to be great at solving the problem whose solution is "this is the person in the county court clerks office you need to talk to."

Re: Claude for Legal

#145
You would think these AI companies would learn by now that the legal field is not a place where you screw around. I mean, we just have to look at the record. Apparently, these particular AI boosters are incapable of learning

Re: Claude for Legal

#146

Earlier quoted context omitted.

[flagged]

I'm not sure if you're joking but there's actually active court cases right now where they have done just that Just a few of the perps: Hisham Abugharbieh (Florida student murders), Jonathan Rinderknecht (Palisades Fire arson), Phoenix Ikner (FSU shooter), Ryan Schaefer (Missouri State vandalism) There's also that thing involving somebody I think he used to be in the NFL and he was using ChatGPT to try to hide the bo…

Interesting these are all examples of people that we want in jail forever. So again this is not gonna happen to you it’s gonna happen to the psychos which is probably what you want

Re: Claude for Legal

#147
post #143
post #7

As a lawyer, I'm excited about this, but there are two roadblocks that I'm not sure how Anthropic will navigate: (1) For non-lawyers who use these skills/connectors/whatchamacallits to try to get legal advice, their communications are not protected by attorney-client privilege. This will absolutely bite some people in the ass. (2) If a lawyer uses this with confidential client information (which, to the uninitiated,…

Wonder if people who decide to represent themselves would be allowed to use this service “live” on the courtroom. Usually people who represent themselves fall flat because they don’t know when it’s appropriate to invoke what and at time appear to putting legal words together that don’t make sense in a given context. This would certainly help these people, if allowed, and they don’t go off the rails.

I'm pretty sure this has already been tried and it... Didn't go well for the person doing it. If memory serves they got a rule 11 sanction out of it.

Re: Claude for Legal

#148

Earlier quoted context omitted.

Indeed, there is no way my terabytes per day of API calls is getting permanently stored anywhere. Perhaps an AI generated summary of it is.

interesting angle - how are/would compressed context (i.e. the parts of the user-LLM transcription likely to be saved) be treated by the courts? Would this be considered hearsay?

It would be treated as evidence, and the defense would be free to argue that it carries very little weight. The judge or jury would be free to decide how meaningful it is.

Re: Claude for Legal

#149

Earlier quoted context omitted.

That might fall under the “unethical” part of my question. Could “probably” get away with it if done carefully, but I’d rather be fully in compliance.

Why would self-hosting for privacy reasons be unethical just because the query would be subject to subpoena in principle?

If you have records, and the discovery is worded in a way that covers the records you have, then deleting or not producing them would be unethical.

More importantly, I’d rather use SOTA models over self-hosted ones if possible.

Post reply on HN