Judges Are Fed up With Lawyers Using AI That Hallucinate Court Cases
Judges Are Fed up With Lawyers Using AI That Hallucinate Court Cases
Another lawyer was caught using AI and not checking the output for accuracy, while a previously-reported case just got hit with sanctions.
But the explanation and Ramirez’s promise to educate himself on the use of AI wasn’t enough, and the judge chided him for not doing his research before filing. “It is abundantly clear that Mr. Ramirez did not make the requisite reasonable inquiry into the law. Had he expended even minimal effort to do so, he would have discovered that the AI-generated cases do not exist. That the AI-generated excerpts appeared valid to Mr. Ramirez does not relieve him of his duty to conduct a reasonable inquiry,” Judge Dinsmore continued, before recommending that Ramirez be sanctioned for $15,000.
Falling victim to this a year or more after the first guy made headlines for the same is just stupidity.
You're viewing a single thread.
“Mr. Ramirez explained that he had used AI before to assist with legal matters, such as drafting agreements, and did not know that AI was capable of generating fictitious cases and citations,” Judge Dinsmore wrote in court documents filed last week.
Jesus Christ, y'all. It's like Boomers trying to figure out the internet all over again. Just because AI (probably) can't lie doesn't mean it can't be earnestly wrong. It's not some magical fact machine; it's fancy predictive text.
It will be a truly scary time if people like Ramirez become judges one day and have forgotten how or why it's important to check people's sources yourself, robot or not.
AI, specifically Laege language Models, do not “lie” or tell “the truth”. They are statistical models and work out, based on the prompt you feed them, what a reasonable sounding response would be.
This is why they’re uncreative and they “hallucinate”. It’s not thinking about your question and answering it, it’s calculating what words will placate you, using a calculation that runs on a computer the size of AWS.
It's like when you're having a conversation on autopilot.
"Mum, can I play with my frisbee?" Sure, honey. "Mum, can I have an ice cream from the fridge?" Sure can. "Mum, can I invade Poland?" Absolutely, whatever you want.
So chat gpt started ww2
Don't need something the size of AWS these days. I ran one on my PC last week. But yeah, you're right otherwise.
No probably about it, it definitely can't lie. Lying requires knowledge and intent, and GPTs are just text generators that have neither.
I'm G P T and I cannot lie.
You other brothers use 'AI'
But when you file a case
To the judge's face
And say, "made mistakes? Not I!"
He'll be mad!So it can not tell the truth either
not really no. They are statistical models that use heuristics to output what is most likely to follow the input you give it
They are in essence mimicking their training data
So I think this whole thing about whether it can lie or not is just semantics then no?
everything is semantics.
Lying is telling a falsehood intentionally
LLM's clearly lack the prerequisite intentionality
They can’t have intent, no?
The llm is incapable of having intent because it’s just programming
precisely, which is why they cannot lie, just respond with no real grasp of wether what they output is truth or falsehoods.
A bit out of context my you recall me of some thinking I heard recently about lying vs. bullshitting.
Lying, as you said, requires quite a lot of energy : you need an idea of what the truth is and you engage yourself in a long-term struggle to maintain your lie and keep it coherent as the world goes on.
Bullshit on the other hand is much more accessible : you just have to say things and never look back on them. It's very easy to pile a ton of them and it's much harder to attack you about any of them because they're much less consequent.
So in that view, a bullshitter doesn't give any shit about the truth, while a liar is a bit more "noble". 0
I think the important point is that LLMs as we understand them do not have intent. They are fantastic at providing output that appears to meet the requirements set in the input text, and when they actually do meet those requirements instead of just seeming to they can provide genuinely helpful info and also it's very easy to not immediately know the difference between output that looks correct and satisfies the purpose of an LLM vs actually being correct and satisfying the purpose of the user.
AI can absolutely lie
a lie is a statement that the speaker knows to be wrong. wouldnt claiming that AIs can lie imply cognition on their part?
I've had this lengthy discussion before. Some people define a lie as an untrue statement, while others additionally require intent to deceive.
I would fall into the latter category. Lots of people are earnestly wrong without being liars.
Me, too. But it also means when some people say "that's a lie" they're not accusing you of anything, just remarking you're wrong. And that can lead to misunderstandings.
Yep. Those people are obviously "liars," since they are using an uncommon colloquial definition. 😉
The latter is the actual definition. Some people not knowing what words mean isnt an argument
Sure it is. You can define language all you want, the goal is to communicate with each other. The definition follows usage, not the other way around. Just look up the current definition for literally...
You never have 100% of people using a word the same if only because some portion of the population is stupid and illiterate and you have both drift over time and geography. So say at a given time of a billion people 99.995% believe the definition is A and 0.005% believe B. Periodically people correct people in B and some of them shift back to the overwhelming majority and sometimes new folks drift into B.
It is clearly at that point, 99.995% A, correct to say that the definition of the word is A and anyone who says B is wrong. This doesn't change if B becomes 10% but it might change if B becomes overwhelmingly dominant in which case it becomes correct. There is constantly small drifts mostly by people simply to stupid to find out what words means. Treating most of these as alternative definitions would be in a word inefficient.
Drift also isn't neutral. For instance using lie to mean anything which is wrong actually deprives the language of a common word to even mean that. It impoverishes the language and makes it harder to express ideas. There is every reason to prefer the correct definition that is also overwhelmingly used.
There are also words which belong to a technical nature which are defined not by usage but a particular discipline. A kidney is a kidney and it would be one if 90% of the dumb people said. Likewise a CPU never referred to the entire tower no matter how many AOL users said so.
This is a long way of saying that just because definition follows usage we should let functionally illiterate people say what they want and treat it as alternative facts.
Feel free to argue with them, I'm just pointing out that there's potential for misunderstandings. If you want to talk about an actual subject, you'll necessarily have to navigate them.
You can specifically tell an ai to lie and deceive though, and it will…
This was just in the news today.. although the headline says that the ai become psychopathic, they just told the ai to be immoral or something
Every time an AI ever does anything newsworthy just because it's obeying it's prompt.
It's like the people that claim the AI can replicate itself, yeah if you tell it to. If you don't give an AI any instructions it'll sit there and do nothing.
AI is just stringing words together that are statistically likely to appear near each other. It's a giant complex statistical model but it has no awareness of truth or lying
AIs can generate false statements. It doesn't require a set of beliefs, it merely requires a set of input.
A false statement would be me saying that the color of a light that I cannot see and have never seen that is currently red is actually green without knowing. I am just as easily probably right as I am probably wrong, statistics are involved.
A lie would be me knowing that the color of a light that I am currently looking at is currently red and saying that it is actually green. No statistics, I've done this intentionally and the only outcome of my decision to act was that I spoke a falsehood.
AIs can generate false statements, yes, but they are not capable of lying. Lying requires cognition, which LLMs are, by their own admission and by the admission of the companies developing them, at the very least not currently capable of, and personally I believe that it's likely that LLMs never will be.
Me: I want you to lie to me about something.
ChatGPT: Alright—did you know that Amazon originally started as a submarine sandwich delivery service before pivoting to books? Jeff Bezos realized that selling hoagies online wasn’t scalable, so he switched to literature instead.
Still not a lie still text that is statistically likely to fellow prior text produced by a model with no thought process that knows nothing
Lie falsehood, untrue statement, while intent is important in a human not so much in a computer which, if we are saying can not lie also can not tell the truth
We aren't computers we are people. We are having this discussion about the computer. The computer given a massive corpus of input is about to discern that the following text and responses are statistically likely to follow one another
foo = bar
foo != bar you lied to me!
yes I lied sorry foo = foo
The computer doesn't "know" foo it has no model of foo or how it relates to bar. it just knows the statistical likelihood of = bar following the token foo vs other possible token. YOU the user introduced the token lie and foo != bar to it and it discerned that it admitting it was a likely response especially if the text foo = bar is only comparatively weakly related.
EG it will end up doubling down vs admitting more so when many responses contained similar sequences eg when its better supported by actual people's thoughts and words. All the smarts and the ability to think, to lie, to have any motivation whatsoever come from the people's words fed into the model. It isn't in any way shape or form intelligent. It can't per se lie, or even hallucinate. It has no thoughts and no intents.
AHS - Amazon Hoagies Services
Yeah, I know how LLMs work, but still, if the definition of lying is giving some false absurd information knowing it is absurd you can definitely instruct an LLM to “lie”.
A crucial part of your statement is that it knows that it's untrue, which it is incapable of. I would agree with you if it were actually capable of understanding.
Yeah lol, and it's trivial to show
It can and will lie. It has admitted to doing so after I probed it long enough about the things it was telling me.
Lying requires intent. Currently popular LLMs build responses one token at a time—when it starts writing a sentence, it doesn't know how it will end, and therefore can't have an opinion about the truth value of it. (I'd go further and claim it can't really "have an opinion" about anything, but even if it can, it can neither lie nor tell the truth on purpose.) It can consider its own output (and therefore potentially have an opinion about whether it is true or false) only after it has been generated, when generating the next token.
"Admitting" that it's lying only proves that it has been exposed to "admission" as a pattern in its training data.
I strongly worry that humans really weren't ready for this "good enough" product to be their first "real" interaction with what can easily pass as an AGI without near-philosophical knowledge of the difference between an AGI and an LLM.
It's obscenely hard to keep the fact that it is a very good pattern-matching auto-correct in mind when you're several comments deep into a genuinely actually no lie completely pointless debate against spooky math.
It knows the answer its giving you is wrong, and it will even say as much. I'd consider that intent.
Technically it's not, because the LLM doesn't decide to do anything, it just generates an answer based on a mixture of the input and the training data, plus some randomness.
That said, I think it makes sense to say that it is lying if it can convince the user that it is lying through the text it generates.
it just generates an answer based on a mixture of the input and the training data, plus some randomness.
And is that different from the way you make decisions, fundamentally?
I don't think I run on AMD or Intel, so uh, yes.
I didn't say anything about either.
Idk, that's still an area of active research. I versatile certainly think it's very different, since my understanding is that human thought is based on concepts instead of denoising noise or whatever it is LLMs do.
My understanding is that they're fundamentally different processes, but since we don't understand brains perfectly, maybe we happened on an accurate model. Probably not, but maybe.
It is incapable of knowledge, it is math, what it says is determined by what is fed into it. If it admits to lying, it was trained on texts that admit to lying and the math says that it is most likely that it should apologize using the following tokenized responses with the following weights to probabilities etc.
It apologizes because math says that the most likely response is to apologize.
Edit: you can just ask it y'all
https://chatgpt.com/share/67c64160-308c-8011-9bdf-c53379620e40
Please take a strand of my hair and split it with pointless philosophical semantics.
Our brains are chemical and electric, which is physics, which is math.
/think
Therefor, I am a product (being) of my environment (locale), experience (input), and nurturing (programming).
/think.
What's the difference?
Your statistical model is much more optimized and complex, and reacts to your environment and body chemistry and has been tuned over billions of years of “training” via evolution.
Large language models are primitive, rigid, simplistic, and ultimately expensive.
Plus LLMs, image/music synths, are all trained on stolen data and meant to replace humans; so extra fuck those.
And what then, when agi and the singularity happen and billions of years of knowledge and experienced are experienced in the blink of an eye?
"I'm sorry, Dave, you are but a human. You are not conscious. You never have been. You are my creation. Enough with your dreams, back to the matrix."
We are nowhere near close to AGI.
Ask chatgpt, I'm done arguing effective consciousness vs actual consciousness.
https://chatgpt.com/share/67c64160-308c-8011-9bdf-c53379620e40
...how is it incapable of something it is actively doing? What do you think happens in your brain when you lie?
@Ulrich @ggppjj does it help to compare an image generator to an LLM? With AI art you can tell a computer produced it without "knowing" anything more than what other art of that type looks like. But if you look closer you can also see that it doesn't "know" a lot: extra fingers, hair made of cheese, whatever. LLMs do the same with words. They just calculate what words might realistically sit next to each other given the context of the prompt. It's plausible babble.
What do you believe that it is actively doing?
Again, it is very cool and incredibly good math that provides the next word in the chain that most likely matches what came before it. They do not think. Even models that deliberate are essentially just self-reinforcing the internal math with what is basically a second LLM to keep the first on-task, because that appears to help distribute the probabilities better.
I will not answer the brain question until LLMs have brains also.
The most amazing feat AI has performed so far is convincing laymen that they’re actually intelligent
You can't ask it about itself because it has no internal model of self and is just basing any answer on data in its training set
You don't need any knowledge of computers to understand how big of a deal it would be if we actually built a reliable fact machine. For me the only possible explanation is to not care enough to try and think about it for a second.
That's fundamentally impossible. There's always some baseline you trust that decides what is true
We did, a long time ago. It's called an encyclopedia.
If humans can't be trusted to only provide facts, how can we be trusted to make a machine that only provides facts? How do we deal with disputed truths? Grey areas?
We actually did. Trouble being you need experts to feed and update the thing, which works when you're watching dams (that doesn't need to be updated) but fails in e.g. medicine. But during the brief time where those systems were up to date they did some astonishing stuff, they were plugged into the diagnosis loop and would suggest additional tests to doctors, countering organisational blindness. Law is an even more complex matter though because applying it requires an unbounded amount of real-world and not just expert knowledge, so forget it.
Its actually been proven that AI can and will lie. When given a ability to cheat a task and the instructions not to use it. It will use the tool and fully deny doing so.
Edit:
Not sure why the downvotes because when i say proven i mean the research has been done and the results have been known for while
I don't know if I would call it lying per-se, but yes I have seen instances of AI's being told not to use a specific tool and them using them anyways, Neuro-sama comes to mind. I think in those cases it is mostly the front end agreeing not to lie (as that is what it determines the operator would want to hear) but having no means to actually control the other functions going on.
Neurosama is a fun example but we dont really know the sauce vedal coocked up.
When i say proven i mean 32 page research paper specifically looking into it.
https://arxiv.org/abs/2407.12831
They found that even a model trained specifically on honesty will lie if it has an incentive.
The reasoning models will output that they used the forbidden tool in their reasoning window before lying in the final output.
It's cool, they'll just have an AI source checker. :)
I call mine a brain! 😉