Verification should never be optional.
this post was submitted on 17 Jul 2026
489 points (99.2% liked)
Fuck AI
7560 readers
1150 users here now
"We did it, Patrick! We made a technological breakthrough!"
A place for all those who loathe AI to discuss things, post articles, and ridicule the AI hype. Proud supporter of working people. And proud booer of SXSW 2024.
AI, in this case, refers to LLMs, GPT technology, and anything listed as "AI" meant to increase market valuations.
founded 2 years ago
MODERATORS
I feel like citing AI hallucinations should lead to disbarment or whatever it's called when you're not allowed to do legal stuff anymore.
It should be the same penalty as deliberately lying on one of these, which I think is an actual crime that can come with prosecution and jail time.
justice slop
Sadly justice slop has been a thing since before AI, it just lowered the bar dramatically. (Pun intended.)
I was listening to a podcast by a lawyer, and how back then, when the DOJ brought a case, it was fully expected for them to win, as their evidence would be detailed, no gaps or unknowns. It also took them a really long time to start/end cases.
Then Donald Trump's magic touch has turned the DOJ into a huge laughing stock. Their cases skyrocketed, many falling apart because of shitty homework and research.
This once respected DOJ by lawyers everywhere is now becoming as bad as ambulance chasers and injury lawyers.
The mission was to deligitimize any/all institutions.
To be fair, this is not unique to the DOJ. Even massive biglaw firms have been reprimanded by courts, several times, for fake AI citations.
God. Does anyone remember the 1995 film The Net?
https://www.imdb.com/title/tt0113957/
When Angela Bennett, a computer programmer, stumbles upon government secrets, she finds herself on the run from an unknown enemy hell-bent on destroying her completely.
Spoilers for a 30 year old movie:
spoiler
Basically there's a shadowy corpo-government organization that maintains power through altering digital records.
Some examples include convincing a conservative Senator to kill himself by altering digital records to convince him he's HIV-positive. Later they target the protagonist by deleting her records from official government databases and attaching her photo and other info to an alias, an imaginary person wanted for a litany of serious felonies.
It seemed outlandish at the time, but that's really where we are headed.
Sigh, this seems to be a crosspost of another feddiverse post sourced from a reddit post of a screenshot of a twitter post, none of which link to the post of the actual source material.
Sorry for not linking to a fascist enabling website that generates ai child pornography
You should learn how to search courtlistener yourself, and recognize the diminished value of a picture of a picture. https://storage.courtlistener.com/recap/gov.uscourts.miwd.120609/gov.uscourts.miwd.120609.11.0.pdf
Thank you for posting the link.
Btw, I can recognize that something is important and hope others will continue the effort. My focus is mainly scraping various sites for crimes of ice, and I have a life outside if that.
Thank you both. Team work makes the meme work.
@BCOVertigo@piefed.world
I've never heard of it, and found the search pretty frustrating.
Here's the case, for anyone else looking. https://www.courtlistener.com/docket/73284734/daghra-v-hinkley/
And the specific document with the judge's response.
https://www.courtlistener.com/docket/73284734/11/daghra-v-hinkley/
That is missing my point, you could have grabbed the attached files of the Twitter post, then attached them to the post.
As it stands now, you haven't rwally done anything bu act like an echochamber, we have no way to independently verify the claims made here.
Look, sourcing of information is important even if you have to use a source you dislike.
If you believe the source to be problematic, you add notes to that effect.
By refusing to provide a source you are simply posting clickbait, clickbait without any substance.
The best thing you can do in these cases is to find an alternative source that you find more credible and use that. And by that I don't just mean finding a post or article referencing the source you dislike, but rather an article that is either investigating the claims, or you investigate them yourself.
I refuse to open an X.com link
Yet you repost their posts, still driving traffic to them.
Show me how it's done then, get the links to the story and show the evidence.
I have other priorities right now like getting my daughter to summer camp
I have no intention of doing your job for you, you made the post, you provide the source.
In the current era of AI fakes, outright lies, misunderstandings, and ignorance clouding the judgement of people, posting accurate sources is important so that everyone has the same opportunity to verify what is being discussed.
Look, my first comment was mainly a joke of the absurdity of the reality of people sending random screenshots of social media posts as fact without bothering to add a direct link as a source. You could just have ignored it or even agreed with me, added a quick source link to the original post, and that would have been fine, but you came in fighting, so I had to respond in kind.
Anyway, put down your phone and help your daughter to get to summer camp!
If my options are posting a screenshot and hope the community follows up, or nothing, then I'd rather post the screenshot. Dispite possible abuse from people like you.
In the future, I will try to remember to ask for help from the community when I don't have the time to research myself.
Abuse?
That wasn't abuse, criticism, yes, abuse, nope.
It's really easy to criticize and mock people.
How many reports of the US regime 's crimes have you reviewed today?
How much time a week do you volunteer to help your community?
I never mocked you, if I made you feel mocked, then I apologize, it was never my intention to do so, my intention was to point out the absurdity of the phenomenon of reposts like it.
As for the other part of your post, that is an entirely different topic than this thread.
We are a community, we can iterate on what each other say.
In the future, I'll try to ask for community help in the post to help research when I do not have the time.
You are not wrong that this should include a link to the doc in the screenshot. But I think you are wrong in how you demanded it. This instagated a emotional response from me, so I'm sorry about that. Also, accusations of posting clickbait and adding to an echo chamber are inappropriate.
- I have no benefit if this post is clicked or upvoted.
- This is really happening, this is legal evidence used to send people to concentration camps or deport them (sometimes to countries they have never been)
Remember, not everything you don't like needs to be accused of something, you can just simply help fix it sometimes, which I guess we are doing now.
if you haven't yet seen, the link to the document was posted in the comments.
https://storage.courtlistener.com/recap/gov.uscourts.miwd.120609/gov.uscourts.miwd.120609.11.0.pdf
Sounds like a good way to continue!
Have a great weekend!
How long do you think until a ridiculous SCOTUS ruling from the right used to remove even more rights and empower the presidents office is citing AI slop?
At this point thanks to LLMs Reddit shitposts probably hold more sway in the government than Trumps gaggle of incompetents and grifters.
Wow, this is lazy. I feel like you could probably set up an AI agent thing to check every case it references to make sure it is both real and relevant. But they were so lazy not only did they not check it themselves, they didn't even have a clanker check it.
I don't think LLMs are great at figuring out if something is written by an LLM, or factual.
I wouldn't have it do that. I'm thinking more along the lines of an AI that drafts legal documents.
- It would find a reference to a case, something an ai is good at.
- Then it would send out a query to the database to check if that case exists, something you would need to build some tooling for.
- If it retrieves anything it would then check if the source and the document agree
- if not, it would run a loop to re-write that section.
I'm not like an agent engineer or anything like that, and I'm sure there's more efficient ways of having an LLM draft legal documents. But so long as you can get it to reference primary sources there should be a lot less cleanup work to do.
You dont need to - you just have to make sure you set up one or two to find holes. Make it 50 points per hole, and 1000 bonus if all is proved wrong. Make two compete and let them know
And then you can have more that check those results for faults!! With only 500 agents working full time you have quite a great setup
Or, hear me out, we don't use generative AI for anything related to human welfare.
Or, hear me out, we don't use generative AI for anything ~~related to human welfare~~.
Asking a chatbot to stop lying is about as useful as asking a river to stop flowing. It won't happen unless you do it yourself.
I lay out how I would do it in another comment. I think it would be pretty easy to get something rough draft quality pretty easy. The difficult problem is compliance, you would probably have to use either local or legal work approved LLMs. But I'm saying they actually just used a chat bot instead of an agent that could give you better results. It's pretty easy to get a clanker not to lie, use it as a tool to rearrange text and don't let it be creative.
The law is a hundreds of billions of dollar annual industry, just in the US. It was worth 367 billion just in 2022. An AI that flawlessly checks over a legal brief would probally be worth billions/year because that means it could also write them flawlessly. Since its so easy, you should probally whip it up and make your billions.
Hell, you seem to have a way to solve chatbots persistent error rate that trillion dollar companies can't fix, so you might actually have a trillion dollar answer. Why stop at just law if "its so easy to make a LLM not lie?"
It's a saturated business, there's dozens of companies already doing it. I don't like AI, I think it's destroying a lot of parts of our life. But it's here and it's only going to sink it's claws deeper. Also I'm not describing chat bots.
And I'm saying that lying is inherent to the technology. Generative AI uses probability to generate output. It's constantly rolling dice when determining output. Any data fed into it is used to calculate odds on what words should follow other words.
It's like making a smoothie and then trying to consume specific vitamins. No matter how good your smoothie making machine is, that's gonna be an impossible task because blending everything together is inherent to making a smoothie. You'd be much better served just eating individual bananas to get your potassium intake.
And I agree with you. When you let an AI be creative it will lie. But if they are reading documents they generally don't. That information is now a part of their context and it will reuse that information. It's a good fuzzy search. It is however terrible at a lot of other things.
Edit: it's like a plagerism machine! Which in this one case is probably good
If you feed a chatbot the sentences, "Bananas are a good source of potassium. Bananas are not a good source of protein." There is a good chance it will spit out the phrase, "Bananas are a good source of protein." because the phrase "good source of protein" is in the original data it was fed. There will always be a significant chance of spitting out this phrase.
No matter how much you improve the AI, because it uses probabilistic generation, it will always be cobbling together sentences without any regard for what the overall sentence means. An intern will eventually learn case law, an AI agent will never be trustworthy. When you ask an AI agent a question, it will always lie to you about being certain of its answer because its answer is built upon a roll of the dice.
For "real", you don't even need an AI. For "relevant", I wouldn't trust an AI. I don't think AI can solve that problem.