Win / KotakuInAction2
KotakuInAction2
Communities Topics Log In Sign Up
Sign In
Hot
All Posts
Settings
All
Profile
Saved
Upvoted
Hidden
Messages

Your Communities

General
AskWin
Funny
Technology
Animals
Sports
Gaming
DIY
Health
Positive
Privacy
News
Changelogs

More Communities

frenworld
OhTwitter
MillionDollarExtreme
NoNewNormal
Ladies
Conspiracies
GreatAwakening
IP2Always
GameDev
ParallelSociety
Privacy Policy
Terms of Service
Content Policy
DEFAULT COMMUNITIES • All General AskWin Funny Technology Animals Sports Gaming DIY Health Positive Privacy
KotakuInAction2 The Official Gamergate Forum
hot new rising top

Sign In or Create an Account

86
Lawyers keep using AI for legal briefs. AI keeps inventing cases to include in the brief. Judge spanks lawyer who got caught out. (twitter.com)
posted 312 days ago by norwegianwikin 312 days ago by norwegianwikin +86 / -0
38 comments share
38 comments share save hide report block hide replies
Comments (38)
sorted by:
▲ 38 ▼
– SR388-SAX 38 points 312 days ago +38 / -0

It's bizarre to me that someone would do this and not at least look up the cases themselves to verify.

Immediate disbarment, IMO.

permalink save report block reply
▲ 24 ▼
– Ahaus667 24 points 312 days ago +24 / -0

To be completely fair the way “precedent” is treated by our legal system they didn’t do much different by completely fabricating things.

permalink parent save report block reply
▲ 16 ▼
– Bouldabassed 16 points 312 days ago +16 / -0

Make up case and judge names? Career over (rightfully).

Make up precedent and completely misrepresent what a given case concluded and use that as justification for making an activistic ruling? All good.

permalink parent save report block reply
▲ 2 ▼
– PartiallyReddited 2 points 311 days ago +2 / -0

Some of the highest paid lawyers in the country have already done that to eat orange Badman. AI is just using the tactics it was trained on.

permalink parent save report block reply
▲ 4 ▼
– tralbolh 4 points 311 days ago +4 / -0

Pro Hac Vice revoked means she was not a member of the bar in the first place. Basically a special guest attorney who will not be returning for another episode.

permalink parent save report block reply
▲ 3 ▼
– milkcowhplus 3 points 311 days ago +3 / -0

I work in a field where I have occasion to use chatgpt to find relevant citations, but yeah, I have two brain cells to rub together, so I actually look up the citations chatgpt spits out to confirm that 1) they exist (sometimes they don’t) and 2) they say what chatgpt says they say (sometimes they don’t).

People who think AI is a full-on replacement for human thinking and judgment, and not a tool to complement it, are retarded.

permalink parent save report block reply
▲ 33 ▼
– TheOutlaw 33 points 312 days ago +33 / -0

Counsel must write a letter to the 3 judges to whom she attributed fake cases

So a diversity hire. How about replacing her with the qualified white male she stole the job from?

permalink save report block reply
▲ 9 ▼
– JustHereForTheSalmon 9 points 312 days ago +9 / -0

He could use ChatGPT to write those three letters for the funniest outcome.

permalink parent save report block reply
▲ 12 ▼
– captain-shitpost 12 points 312 days ago +12 / -0

Ask chatgpt on the topic of immigration law or discrimination against black people...

It massively pushes open border interpretations that just aren't there in the law. Situations where x happens, and the exception y is less than 1%, it will use language like

x may happen, but often does not. instead, most people y.

As for discrimination cases, it literally makes shit up to argue that it's likely or "may be" discrimination... until you swap in white people, christians, straight people. Then suddenly disparate impacts doesn't apply, and they use the racist term "reverse discrimination" which argues that racism can only go from certain races to certain others, which is definitionally racist in itself.

permalink save report block reply
▲ 6 ▼
– exarch 6 points 312 days ago +6 / -0

That’s what happens when you train it on Reddit and reinforce any leftist tendencies it may have.

I remember when Microsoft Tay happened and people on the right (/pol/ for example) were circle jerking saying that AI will always be right wing because our positions aren’t bullshit. I tried to tell people that AI will be whatever you tell it to be. I said that silicon valley will most definitely create an SJW AI and I was right.

At least now we can see it clearly and react to it instead of deluding ourselves

permalink parent save report block reply
▲ 3 ▼
– RoulerBleu 3 points 311 days ago +3 / -0

An AI notices patterns. Noticing patterns is racist, sexist, homophobic, transphobic, xenophobic, and almost every ''istism''.

In order to get an ''acceptable'' AI, leftists have to lobotomize it and feed it carefully-curated selective information + outright lies, and then prevent outside information from tainting it with input from reality.

permalink parent save report block reply
▲ 2 ▼
– captain-shitpost 2 points 312 days ago +2 / -0

yup. there are plenty of other topics where it's pure bullshit, like house seats apportioned because of illegal aliens, to gerrymandering stats, to the current DC crime scandal.

spoiler alert: the top 5 most gerrymandered states are all stealing about 20-30 seats in the house total to the dems.

if you argue with it, provide something contradictory, it'll admit you're right, but try to deflect/diminish, and if you keep pushing, it'll fold and admit you're right. it's insane because people give it this impression of being unbiased.

permalink parent save report block reply
▲ 9 ▼
– MattTheBlack 9 points 312 days ago +9 / -0

Absolutely retarded. AI is only as good as the information provided and it's Absolutely shit at finding legal information. It should only be used to spell and grammar check

permalink save report block reply
▲ 8 ▼
– deleted 8 points 312 days ago +8 / -0
▲ 5 ▼
– Yashimata 5 points 312 days ago +5 / -0

It still has its limits. If you tell it to hallucinate often enough you'll start seeing patterns. I don't know what Grok's patterns are, but I imagine ChatGPT just shits out character after minority character who are either lawful (if the setting is "good") or chaotic (if they setting is "evil") good, who are hell-bent on fighting injustices and promoting diversity. Let the orcs raise your children, bigot.

permalink parent save report block reply
▲ 3 ▼
– auroch 3 points 312 days ago +3 / -0

I use GPT to brainstorm my Shadowrun campaign, which is delightfully meta in my opinion. Generates a fair bit of stuff I don't use, but that's to be expected. Not overly woke in this context either, so that's nice.

I do wish it would stop going, "great idea! This will create amazing dramatic tension when your party has to choose between what's right and what's good for them!" I'm always like, ".... my party is a tabletop RPG party. They don't care at all about what's right." I love them to pieces, though :)

permalink parent save report block reply
▲ 4 ▼
– DemolitionsPanda 4 points 312 days ago +4 / -0

Shadowrun is amazing, though I strongly feel that the rules went down-hill after 3rd edition with the disconnect between stats and skill advancement.

I played a campaign from the early 1st edition adventures all the way through Bug City and Super Tuesday to the Renraku Archology Shutdown.

It was a blast.

permalink parent save report block reply
▲ 3 ▼
– auroch 3 points 312 days ago +3 / -0

Far and away my favorite tabletop setting. The juxtaposition of fantasy, magic, and cyberpunk corporatism is so good. I believe I heard Catalyst has gone at least somewhat woke, but for anyone interested, you should be able to find older used books without giving them a dime.

permalink parent save report block reply
▲ 2 ▼
– MattTheBlack 2 points 312 days ago +2 / -0

I meant for legal work

permalink parent save report block reply
▲ 1 ▼
– deleted 1 point 311 days ago +1 / -0
▲ 3 ▼
– akira2501 3 points 312 days ago +3 / -0

The information has to be notated by a human. It can't "find" anything.

They've gotten lazy and decided to use "reinforcement learning" which means one LLM notates information for another LLM. That's why it's gotten recursively worse.

It's a polluted system. It cannot possibly last. The only reason it trudges on is because Google and Microsoft foolishly went a billion or three long on this "tech."

permalink parent save report block reply
▲ 2 ▼
– MargarineMongoose 2 points 311 days ago +2 / -0

It should only be used to spell and grammar check

Even this is a dubious use of it, as it can still fuck that up. It's just statistically far less likely to.

Besides, what legal office doesn't have Microsoft Word already?

permalink parent save report block reply
▲ 7 ▼
– Adamrises 7 points 312 days ago +7 / -0

The amount of people across all industries and academia that have become entirely dependent on AI answers to all their problems in such a short fucking time frame is alarming in a way that I don't think enough fear is being given.

Like, the absolute meltdown over the ChatGPT5 change and the amount of people openly admitting that it is their entire social life, half their professional one, and more is wild. And sure those are the loser dregs on places like reddit, but its clearly becoming a problem rapidly as shown by things like this.

permalink save report block reply
▲ 4 ▼
– MargarineMongoose 4 points 311 days ago +4 / -0

I think it really just reveals the competency crisis that already existed. It was just easier to ignore when the missteps weren't so flagrant or still had some veneer of plausible deniability. Now it's just full on "everyone is retarded and it shows."

Same deal for the people making it their social life. It reveals just how sick society is. That people are using it as a substitute for meaningful human relationships means that society is no longer facilitating healthy human interaction to a degree that's even worse than we were willing to admit to ourselves.

We're pulling back the curtain and realizing just how bad everything actually has been this whole time.

permalink parent save report block reply
▲ 4 ▼
– tralbolh 4 points 311 days ago +4 / -0

It is partially that the traditional search engines have gotten progressively worse year over year.

A time machine, even if just limited to being able to query a google server from 17 years ago, would be an extremely useful thing.

permalink parent save report block reply
▲ 3 ▼
– Adamrises 3 points 311 days ago +3 / -0

Agreed on that. Shit Google's AI summary nonsense takes up a giant chunk of the screen on anything you search anyway. And then the first 10 pages are the same generic sites that tell you a cursory summary or "consensus" based information.

That is one of the ways you can smartly use the AI for these purposes. Rough drafts and hopping off points. Like, ask about your case and then look up what it finds to do the real research.

Its what having a calculator with internet on your phone did for math. It negates the need to memorize literally every random equation, and instead lets you focus on applying it. But like so many did with that, they relied on it so heavily they can't even understand what those working parts do and become impotent without their cheatsheets.

permalink parent save report block reply
▲ 1 ▼
– HallucinatoryBeing 1 point 312 days ago +1 / -0

@grok is this true?

permalink parent save report block reply
▲ 5 ▼
– Adamrises 5 points 312 days ago +5 / -0

I can't believe we have reached the point where

Chat is this real

was the preferable alternative.

permalink parent save report block reply
▲ 3 ▼
– RoulerBleu 3 points 311 days ago +3 / -0

I have an ''online friend'' who basically asks Grok and chatGPT everything.

And then argues in favor of the bot's answer when I tell him how the bot is incorrect. Even if I get him to admit the real answer isn't what the ''AI'' says it is, he will default back to it within the same conversation.

permalink parent save report block reply
▲ 6 ▼
– Sneak_King 6 points 312 days ago +6 / -0

I like that the judge specifically describes them as hallucinatory cases. That's a decent understanding of LLMs.

permalink save report block reply
▲ 6 ▼
– ParadigmShift2070 6 points 312 days ago +6 / -0

It's wild that "don't cite nonexistent sources" is not the default, instead, you have to specify you don't want fakes

permalink save report block reply
▲ 6 ▼
– Mpetey123 6 points 312 days ago +6 / -0

8 years of expensive schooling and he depends on an app. It's a choice between a decade of higher education or asking my smartphone for help

permalink save report block reply
▲ 19 ▼
– HallucinatoryBeing 19 points 312 days ago +19 / -0

*she

permalink parent save report block reply
▲ 9 ▼
– akira2501 9 points 312 days ago +9 / -0

All you have to do is think of what the hourly rate for their services is. If they don't have to do the work, can get a machine to do it, then bill for that time anyways, well... the writing is on the wall isn't it?

The judge is being too soft. This is about money. This could be construed as a fraud against her client. You have to hit them in the wallet or they will do it again.

permalink parent save report block reply
▲ 5 ▼
– Benevolentdictator 5 points 312 days ago +5 / -0

Exactly paralegals were writing the briefs for partners before AI came along.

permalink parent save report block reply
▲ 4 ▼
– HOGCRU 4 points 311 days ago +4 / -0

don't worry, she cheated her whole way through school with chatgpt too

permalink parent save report block reply
▲ 2 ▼
– Mpetey123 2 points 311 days ago +2 / -0

She probably did

permalink parent save report block reply
▲ 4 ▼
– tralbolh 4 points 311 days ago +4 / -0

Counsel must write a letter to the 3 judges to whom she attributed fake cases

ChatGPT must write a letter to the 3 judges to whom she attributed fake cases

permalink save report block reply
▲ 2 ▼
– deleted 2 points 312 days ago +2 / -0
▲ 1 ▼
– deleted 1 point 311 days ago +1 / -0

Original 8chan Links to Gamer Gate:

.

The main GG discussion is on the videogames board: https://8chan.moe/v/

.

GamerGate archive is at https://8chan.moe/gamergatehq/

.

GamerGate Wiki:

https://ggwiki.deepfreeze.it/index.php/Main_Page

. . . . . .

. . . . . .

Rules:

.

ONE: Do not advocate for illegal violence or post other illegal activity. (Be aware of your local laws.)

.

TWO: Don't threaten, harass, or impersonate users. Also: don't be a psycho. New users will be held to a higher standard.

.

THREE: Do not post porn.

.

FOUR: NSFW/NSFL content must be flaired NSFW.

.

FIVE: No vote manipulation. Do not break communities.win's features.

.

SIX: No spam or reposts. Do not make more than 5 threads a day.

.

SEVEN: Do not post falsehoods and hoaxes that are obvious to an uncontroversial degree.

. . . . . .

. . . . . .

Moderation Logs:

.

(Two different versions, Scored has more features and is cleaner, but .win let's you see a few more details in certain instances.)

  • Scored
  • .win

Moderators

  • DomitiusOfMassilia
  • C
  • BandageBandolier
  • CarmenOfSandiego
  • The_Shadow_of_Intent
  • SocraticMethod1
  • Kienan
  • Smith1980
Message the Moderators

Terms of Service | Privacy Policy

2026.02.01 - bh6wd (status)

Copyright © 2026.

Terms of Service | Privacy Policy