Just ask ChatGPT what it thinks for some non-existing product and it will start hallucinating.
This is a known issue of LLMs and DL in general as their reasoning is a black box for scientists.
It’s not that their reasoning is a black box. It’s that they do not have reasoning! They just guess what the next word in the sentence is likely to be.
I mean it’s a bit more complicated than that, but at its core, yes, this is correct. Highly recommend this video.
it’s not even a little bit more complicated than that. They are literally trained to predict the next token given a series of previous tokens. The way that they do that is very complicated and the amount of data they are trained on is huge. That’s why they have to give correct information sometimes to sound plausible. Providing accurate information is literally a side effect of the actual thing they are trained to do.
The technology has to follow the legal requirements, not the other way around.
That should be obvious to everyone that’s not an evangelist.
And by the time the system can actually research the facts, the internet is so full of LLM generated nonsense neither human or AI can verify the data.
If scientists made AI, then it wouldn’t be an issue for AI to say “I don’t know”.
But capitalists are making it, and the last thing you want is it to tell an investor “I don’t know”. So you tell it to make up bullshit instead, and hope the investor believes it.
It’s a terrible fucking way to go about things, but this is America…
It’s got nothing to do with capitalism. It’s fundamentally a matter of people using it for things it’s not actually good at, because ultimately it’s just statistics. The words generated are based on a probability distribution derived from its (huge) training dataset. It has no understanding or knowledge. It’s mimicry.
It’s why it’s incredibly stupid to try using it for the things people are trying to use it for, like as a source of information. It’s a model of language, yet people act like it has actual insight or understanding.
Imagine searching your computer for a PDF named “W2.2026”…
Would you rather the computer tell you it’s not in the database? Or would you prefer a random PDF displayed with the title “W2.2026”?
This isn’t a new problem.
You’re getting hung up on “know” instead “has relevant information in it’s database and can access it”.
But besides all that and the other things you got wrong:
It’s still about capitalism for the reasons I just said
You don’t seem to understand. There is no database.
You do not understand how these things actually work. I mean, fair enough, most people don’t. But it’s a bit foolhardy to propose changes to how something works without understanding how it works now.
There is no “database”. That’s a fundamental misunderstanding of the technology. It is entirely impossible to query a model to determine if something is “present” or not (the question doesn’t even make sense in that context).
A model is, to greatly simplify things, a function (like in math) that will compute a response based on the input given. What this computation does is entirely opaque (including to the creators). It’s what we we call a “black box”. In order to create said function, we start from a completely random mapping of inputs to outputs (we’ll call them weights from now on) as well as training data, iteratively feed training data to this function and measure how close its output is to what we expect, adjusting the weights (which are just numbers) based on how close it is. This is a gross simplification of the complexity involved (and doesn’t even touch on the structure of the model’s network itself), but it should give you a good idea.
It’s applied statistics: we’re effectively creating a probability distribution over natural language itself, where we predict the next word based on how frequently we’ve seen words in a particular arrangement. This is old technology (dates back to the 90s) that has hit the mainstream due to increases in computing power (training models is very computationally expensive) and massive increases in the size of dataset used in training.
Source: senior software engineer with a computer science degree and multiple graduate-level courses on natural language processing and deep learning
Btw, I have serious issues with both capitalism itself and machine learning as it is applied by corporations, so don’t take what I’m saying to mean that I’m in any way an apologist for them. But it’s important to direct our criticisms of the system as precisely as possible.
It’s not a database. God, how many years is it going to take before people understand just what LLMs are and are not capable of?
It is made by scientists. The problem is that said scientists are paid by investors mostly, or by grants that come from investors.
Just put this into GPT 4.
What’s your view of the fizbang Raspberry blasters?
Gpt ‘I’m not familiar with “fizbang Raspberry blasters.” Could you provide more details or clarify what they are?’
It’s a drink making machine from china
Gpt ‘I don’t have any specific information on the “fizbang Raspberry blasters” drink making machine. If it’s a new or niche product, details might be limited online.’
So, in this instance is didn’t hallucinate, i tried a few more made up things and it’s consistent in saying it doesn’t know of these.
Explanations?
Chatgpt and gpt4 are two different things. Gpt4 is like the engine and chatgpt is like a car. In early version they were pretty much the same thing, but nowadays they have implemented so much in chatgpt.
On top of that chatgpt4 is constantly trained for these scenarios, it is no longer a base model.
Oh ok thanks i thought this thread was about AI LLMs in general.
Weird i was downvoted for demonstrating the very thing that apparently (according to these very learned comments) AI can’t do, actually doing it well. Seems like irrational bubble hate to me, common on reddit but getting more so on Lemmy it seems. “that guys asking topic based questions that make our comments look poorly thought out and potentially wrong, burn him”
This is a thing that is true of all LLMs, but it seems like you’re misunderstanding the core issue. It CAN give outputs like that sometimes. What we CAN’T do is force it to give outputs like that ALL the time.
It will answer “I don’t know” if its predictive text model guesses that the most common response to this would be “I don’t know”. To do that, to simplify a little, you could imagine that it reads your question, compares that to all the text in its training data, and tries to find the conversation that looks most like the question you asked, then answers whatever the person in the training data answered. But your exact question wasn’t in its training data, so if you took that mental model, and instead had it compare to 1000 similar looking things in its training model and average them, then it would hopefully do a better job of coming up with something at least close to what you actually asked. Now take it to a million, or a billion.
When we’re asking questions about the real world, we would prefer for it to answer based on knowledge about the real world. But what if it “matches” data from a work of fiction? Or just someone who doesn’t know what they’re talking about? Or true information, but about a different subject?
It doesn’t know anything. It doesn’t understand anything you say. It just looks at patterns that it learned from the training data and tries to guess what words are most likely to be said in that case. In other words, “here’s one case where it didn’t hallucinate” and “it will never hallucinate” are not the same thing at all.
Edit: To clarify, it doesn’t search its training data to answer your question, so asking “was this in the training data” is impossible. By the time you interact with it, the data is long gone. It was just used for training.
Ok very long and detailed response, i was responding to the initial comments that explicitly said if you give ai a made up thing it will definitely hallucinate. Which i demonstrated to be false in (multiple times). I’m not suggesting it doesn’t hallucinate a lot of the time still, but the comments were making out its 100% broken, and it clearly works for many queries very effectively, despite its limited applications. Im just suggesting we don’t throw the baby out with the bathwater.
I think the trouble is, what baby are we throwing out with the bathwater in this case? We can’t prevent LLMs from hallucinating (but we can mitigate it somewhat with carefully constructed prompts). So, use cases where we’re okay with that are fair game, but any use case (or in this case, law?) that would require the LLM never hallucinates aren’t attainable, and to get back earlier, this particular problem has nothing to do with capitalism.
It is made by scientists. And we don’t know how to make the model determine whether or not it knows something. So far, we only have tools that tell us that something probably wasn’t in the training set (e.g. using variance across models in a mixture of experts setup), but that doesn’t tell us anything about how correct it is.
Uh, I understand the sentiment, but the model doesn’t know anything. And it’s legit really hard to differentiate between factual things and random bullshit it made up.
Yeah, no one can make it say “I don’t know” because it is not really AI. Business bros decided to call it that and everyone smiled and nodded. LLMs are 1 small component (maybe) of AI. Maybe 1/80th of a true AI or AGI.
Honestly the most impressive part of LLMs is the tokenizer that breaks down the request, not the predictive text button masher that comes up with the response.
Honestly the most impressive part of LLMs is the tokenizer that breaks down the request, not the predictive text button masher that comes up with the response.
Yes, exactly! It’s ability to parse the input is incredible. It’s the thing that has that “wow” factor, and it feels downright magical.
Unfortunately, that also makes people intuitively trust its output.
Was gonna say, the AI doesn’t make up or admit bullshit, its just a very advanced a prediction algorithm. It responds with what the combination of words that is most likely the expected answer.
Wether that is accurate or not is part of training it but you’ll never get 100% accuracy to any query
If it can name what the most likely combination is, couldn’t it also know how likely that combination of words is?
It’s not actually deciding anything, the AI thinking is marketing fluff really. But yes that’s called confidence rating and it does. But at the scale of something like chatgpt that uses a snapshot of the entire internet and is non mutable there’s no way to train it for every possible question. If you ask about a topic 99% of the internet gets wrong it’ll respond the wrong thing with 99% confidence
No, because that requires it to understand the words. It doesn’t.
If it has been trained using questionable sources, or if it’s training data includes sarcastic responses (without understanding that context), it isn’t hard to imagine how confidently wrong some of the responses could be.
It “knows” as in it has access to the information and the ability to provide the right info for the right context.
Any part of that process the AI can just “bullshit” and fills in the gaps with random stuff.
Which is what you want when it’s “learning”. You want it to try so it’s attempt can be rated, and the relevant info added to its “knowledge”.
But when consumers are using it, you want it to say “I can’t answer that”. But consumers are usually stupid and will buy/use the one that says “I can’t answer that” the least.
And it’s legit really hard to differentiate between factual things and random bullshit it made up.
Which is why AI should tell end users “I don’t know” more often.
Which is why AI should tell end users “I don’t know” more often.
If you feel this is a simple solution, I strongly suggest you write up exactly how you do this and make yourself a billion dollars.
It “knows” as in it has access to the information and the ability to provide the right info for the right context.
It doesn’t, though, any more than you have access to the information in a pile of 10 million shredded documents.
Right, in this case that we’re talking about…
Do you not understand how “answer unavailable” is a better answer than taking a small percent of strips of paper at random and filling in the rest with words that sound relevant?
It’s like a mad libs
That is what LLMs do in EVERY conversation. Most of the time you don’t notice it, because it fits your expectations.
You know that answer unavailable is better because you have real intelligence, an LLM is just some mathematical functions so it can’t do that. If it could it would be getting much closer to actually being AI.
taking a small percent of strips of paper at random and filling in the rest with words that sound relevant?
It’s like a mad libs
Right. They’re text generators. That’s the technology. It can’t do what you’re demanding because that’s not how it works. LLMs aren’t magic answer machines. They don’t know when to say “answer not available”. They don’t know what they’re being asked. They don’t know anything.
This has nothing to do with scientists vs capitalists and everything with the fact that this is not actually “AI”. Someone called it T9 (word prediction) on steroids and I find that much more fitting with how those LLMs work. It just mimics the way humans talk, but it doesn’t actually converse intelligently or actually understands context - it just looks like it does, but only if you take it at face value and don’t look deeper into it.
We don’t even know how real human intelligence/consciousness works…
Obviously modern chatbots aren’t true AI.
But people call a car with an automatic transmission “an automatic”, no one buts in to explain how the entire car isn’t automatic because you still have to use the steering wheel and pedals.
You’re just being too pedantic, and after a couple other people already were about the exact same thing homie…
Why would I call the car automatic if it has an automatic transmission? Your analogy makes zero sense. Your username definitely does not check out either.Maybe stop being a salty snowflake after being called out for not having a clue of what you’re talking about.
Why would I call the car automatic if it has an automatic transmission?
Because you speak English…
Have a nice life tho, anyone that unironically used “snowflake” and gets that defensive of being wrong is worth blocking.
You make even less sense now. lol
anyone that unironically used “snowflake” and gets that defensive of being wrong is worth blocking.
I suppose you’re suggesting that I block you now, since you got so defensive of being wrong. But if I was that type of guy I would’ve already blocked you for being a literal hamas / terrorist apologist / sympathizer.
It’s just short for automatic transmission, opposed to manual transmission. I think Americans call manual cars sticks though. But they’re not sticks, because sticks are wood and cars are almost always metal. Not metal like the music though.
Edit - thinking on it you could play metal through the car stereo though.
I know the difference between an automatic & manual car & transmission. The analogy just doesn’t make sense, because when you say “automatic / manual car” you’re still referring to something within the car, the transmission system - you’re not actually calling the car to be “automated” or whatever. Calling LLMs “AI” however is nothing but a misnomer and that analogy simply does not compare at all.
There we go. Now that people have calmed their proverbial tits about these thinking machines, we can start talking maturely about the strengths and limitation of the LLM implementations and find their niche in our tools arsenal.
I can’t wait until the AI bubble finally pops.
I’ve got bad news for you though: there will be another new bubble almost immediately. There’s a whole industry based around tech hype cycles and they are constantly throwing shit at the wall to see what sticks. Eventually something will when there’s space for it. It will be just as insufferable as LLMs are, and crypto was before that, and… I actually forget what was before that. Uber? You won’t be able to escape it, because it will dominate the attention economy.
No surprise, and this is going to happen to everybody who uses neural net models for production. You just don’t know where your data is, and therefore it is unbelievably hard to change data.
So, if you have legal obligations to know it, or to delete some data, then you are deep in the mud.
I think of ChatGPT as a “text generator”, similar to how Dall-E is an “image generator”.
If I were openai, I would post a fictitious person disclaimer at the bottom of the page and hold the user responsible for what the model does. Nobody holds Adobe responsible when someone uses Photoshop.I would post a fictitious person disclaimer
… or you could read the GDPR and learn that such excuses are void.
You just wasted a lot of my time. What did I do to deserve this?
LLMs don’t actually store any of their training data, though. And any data being held in context is easily accessible and can be wiped or edited to remove personal data as necessary.
LLMs don’t actually store any of their training data,
Data protection law covers all kinds of data processing.
For example, input is processing, too. Output is processing, too. Section 4 of the GDPR.
If you really want to rely on excuses, you would need wayyy better ones.
Right, so keep personal data out of the training set and use it only in the easily readable and editable context. It’ll still “hallucinate” details about people if you ask it for details about people, but those people are fictitious.
Asking chatgpt for information is like asking for accurate reports from bards and minstrels. Sure, sometimes it fits, but most of it is random stuff stitched together to sound good.
Stop asking a language model for accurate information and problem solved. ChatGPT is not supposed to be a knowledge bank, that’s purely incidental for the amount of training data.
Stop asking a language model for accurate information and problem solved
Hey chatgpt, when did jol’s wife get pregnant and by whom?
/s
Unless they used that bitche’s only fans in the training data, it will definitely not know that.
It doesn’t need to know the real answer to produce a confident sounding answer
And if that answer contains Elon Musk, the world is going to believe it no matter what.
