The executive branch makes all kind of shit up. Why aren't you letting these lawyers? Seems unpatriotic.
this post was submitted on 05 Jun 2026
263 points (98.5% liked)
Technology
85155 readers
5971 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related news or articles.
- Be excellent to each other!
- Mod approved content bots can post up to 10 articles per day.
- Threads asking for personal tech support may be deleted.
- Politics threads may be removed.
- No memes allowed as posts, OK to post as comments.
- Only approved bots from the list below, this includes using AI responses and summaries. To ask if your bot can be added please contact a mod.
- Check for duplicates before posting, duplicates may be removed
- Accounts 7 days and younger will have their posts automatically removed.
Approved Bots
founded 3 years ago
MODERATORS
Should be immediately disbarred
Judges: expressing mild confusion and concern
Headlines: Judges RIP INTO lawyers with their BARE TEETH as they can BARELY CONTAIN their EXPLOSIVE FIREY RAGE!!!!
To be fair, mild confusion and concern is the judicial equivalent. Even the diehard maga judges still behave with decorum, even if their ultimate judgments are unhinged.
There's a common and expected way to do lawyering. If a judge "expresses concern" towards a lawyer, that's a big sign that you dun fucked up.
remember kids GIGO "Garbage In, Garbage Out".
the general public just isn't aware of this. they assume that anything and everything an LLM spits out is fact. Again, the LLM/AI MUST provide a solution. any solution. that solution doesn't need to be correct or even exist but in order to retain users it will and will consistently and constantly provide a hallucinated solution.
And in most cases, without their knowledge, the user is also to blame. I can see a lawyer asking an LLM "are there any cases where such and such happened" trying to find something INCREDIBLY specific to what they need and the LLM 8 times out of 10 will say "yes! there is!" and hallucinate a case that the lawyer needs. You see it all the time if you ask Claude with assistance in trouble shooting a specific tech or code problem. "I'm having issues with this and this, can this library or app or program solve this for me?" "Yes! here's how" and then will tell you something the library or program or whatever can't even do but you've lead it to that direction and it MUST provide a positive retention-able solution.
Even using AI as it SHOULD be used is no longer good. just using it as a sort of rubber duck doesn't help anymore because of GIGO. the training data, across ALL LLMs, is so tainted now you might as well take all the money we're wasting on data centers and tokens and just toss it at the crazy dude outside the bus station and just ask them the same questions.
Socrates was famously called the 'gadfly' because he would pester elites by challenging their supposed wisdom. People that would be highly knowledgeable craftsman, financiers, politicians, etc. would then over estimate their knowledge outside of their realm of expertise and claim to be wise in all things. They would know just enough to sound knowledgeable, but really know very little about what they were talking about (related to the Dunning-Kreuger effect). The takeaway of Socrates' gadfly work is that sometimes wisdom is acknowledging when you do not know something.
AI can have a lot of "knowledge" (it doesn't actually know anything itself, but rather has pattern training and access to data references). But what it doesn't have.... like at all... is wisdom. It doesn't understand, literally, anything at all. It might have some guard rails up to limit is hallucinations. It might even say, sometimes, that it doesn't know. But it is just as happy to ramble on with pure and complete nonsense that is all just a stream of patterns training and probabilistic guesswork. If it has a nugget of data or even an entire library related to your prompt, it will present it to you. But it can 100% just fill any gaps or take the response into tangents that are nothing but guess-the-next-word, probability, looks-good-to-me word salad that can contain several "facts" that are nothing but random word associations from its training with no reference for its basis.
It is a glorified autocorrect. You cannot trust its information blindly, and using it as a tool in this way, especially while staking your fucking law career on it, is goddamned moronic.
That's what happens when you train llm models on reddit.
Good. Get the lawyers irritated about the unapologetic use of AI. Maybe their congressional pals will listen to them
Congress isn't going to do shit.
I'm sure AI could solve this problem. /s
Don’t make mistakes! This is for my really important lawyer job and I could get in trouble!
Make no misteak!
Absolutely, here's a recipe for bistek