Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

"Hallucinating" AIs sound creative, but let's not celebrate being wrong

thereader.mitpress.mit.edu

Post reply on HN
Print view View original

91–100 of 196 posts

Re: "Hallucinating" AIs sound creative, but let's not celebrate being wrong

#91

Post by OrsonSmelles »

mjr00 wrote: post #34
LeoPanthera wrote: post #22

This is an odd article. To me, it seems like the "creative" arts are an ideal arena for AI. After all, there's no such thing as "wrong" art. The article says "well, sometimes what it makes is bad". Well big deal. A lot of human-created art is awful too.

> This is an odd article. To me, it seems like the "creative" arts are an ideal arena for AI. After all, there's no such thing as "wrong" art. Yeah but who wants to consume art purely generated by AI (that is, not human-created with AI support)? Most art sites have had blanket bans, or at least required tagging, on ai-generated art because people hate it so much. Or to put it another way: why are you in the comment s…

>why are you in the comment section of Hacker News and not just asking ChatGPT to generate social media comments on the article?

Because I know the brains generating these comments have rich, diverse experience and a large set of refined, domain-specific heuristics derived from it - that is, they have top-notch training and alignment, at the cost of 20+ years of upfront incubation in an evolution-optimized meat harness with unpredictable success rates (and occasional personality defects). I can encounter new perspectives here that I don't think any current LLM could imitate efficiently or reliably. But if I wanted to know what Fox/MSNBC/NPR commenters had to say about this topic, I would absolutely ask ChatGPT, because those are commodity-grade opinions, and it excels at producing them. (It is not obvious to me that HN will still be an exception for, say, GPT-10 or whatever.)

This is definitely speculative and a little catty, but I suspect a lot of "hatred" of AI art is a) emotional solidarity with working artists who feel economically threatened by it, and b) a personal sense of insecurity along the lines of "what if I love this piece and it turns out to be AI - does that make me a boring NPC/a chump? better reject it as fiercely as possible to avoid introspection!"

Re: "Hallucinating" AIs sound creative, but let's not celebrate being wrong

#92

Post by simplify »

Retric wrote: post #66
simplify wrote: post #52

Earlier quoted context omitted.

No, science is indeed testing random ideas. Limiting what is tested to an arbitrary plausibility score is something funders and sponsors do, not scientists. What you're talking about is ethics, which is important too.

Random ideas include things like: The speed of sound suddenly doubles between 627.500847 to 627.500848 atmospheres. I’ve said it, now try and find someone willing to test it. That’s hard because Scientists want stuff that’s worth spending their limited time on earth looking into.

I assure you, there are many scientists that would love to test that, given the resources to do so.

Of course no one would fund that, meaning it's the funders deciding on what they think is plausible.

Re: "Hallucinating" AIs sound creative, but let's not celebrate being wrong

#94

Post by mjr00 »

OrsonSmelles wrote: post #91
mjr00 wrote: post #34

Earlier quoted context omitted.

> This is an odd article. To me, it seems like the "creative" arts are an ideal arena for AI. After all, there's no such thing as "wrong" art. Yeah but who wants to consume art purely generated by AI (that is, not human-created with AI support)? Most art sites have had blanket bans, or at least required tagging, on ai-generated art because people hate it so much. Or to put it another way: why are you in the comment s…

>why are you in the comment section of Hacker News and not just asking ChatGPT to generate social media comments on the article? Because I know the brains generating these comments have rich, diverse experience and a large set of refined, domain-specific heuristics derived from it - that is, they have top-notch training and alignment, at the cost of 20+ years of upfront incubation in an evolution-optimized meat harne…

> This is definitely speculative and a little catty, but I suspect a lot of "hatred" of AI art is a) emotional solidarity with working artists who feel economically threatened by it, and b) a personal sense of insecurity along the lines of "what if I love this piece and it turns out to be AI - does that make me a boring NPC/a chump? better reject it as fiercely as possible to avoid introspection!"

Strongly disagree, just take a look at the AI-generated tag on pixiv[0] -- it's 99% same-looking garbage, and since there's no barrier to generating, uploaders flood pixiv with tons of images. Browsing tags is basically unusable without it being filtered out.

[0] - https://www.pixiv.net/en/tags/AI-generated/artworks mildly nsfw if you're not logged in, very nsfw if you are. (edit: note that this isn't a lot because pixiv later added a meta-tag for AI art so it's possible for users to filter without pixiv premium; that's how much it was hated.)

Re: "Hallucinating" AIs sound creative, but let's not celebrate being wrong

#95

Post by pdonis »

famouswaffles wrote: post #87
pdonis wrote: post #64

Earlier quoted context omitted.

> humans aren't "optimized" to be correct either Humans at least have a concept of "being correct", even if we don't always set that as our primary goal when communicating. LLMs don't even have a concept of "being correct". That would require having a concept of an "external world" that text refers to, which LLMs don't have. All they have is the text in their training data.

>LLMs don't even have a concept of "being correct" They clearly do as plenty research indicates. You've just decided not to accept this. Pure confirmation bias in action.

> They clearly do as plenty research indicates.

I have seen plenty of researchers claim this. What I have not seen is actual support for such a claim.

Re: "Hallucinating" AIs sound creative, but let's not celebrate being wrong

#96

Post by pdonis »

famouswaffles wrote: post #86
pdonis wrote: post #61

Earlier quoted context omitted.

> Language models (the computation) "care" about making "correct" predictions Predictions of what? LLMs aren't predicting anything. They are, as the GP says, generating text that appears to be fluent and coherent (as long as we don't inquire into actual semantic relationships with anything else) in the context of the prompt.

They are quite literally trained to predict a distribution that could supply the next token given previous context. This is their primary objective.

> They are quite literally trained to predict a distribution that could supply the next token given previous context.

"Previous context" here means text. It does not mean "the actual world". Big difference.

Re: "Hallucinating" AIs sound creative, but let's not celebrate being wrong

#97

Post by SirMaster »

famouswaffles wrote: post #54
farnsworth wrote: post #25

Earlier quoted context omitted.

That's not a useful description. The AI doesn't "know" what it "knows". It's not even filling some gap in its knowledge. It's just putting words together that statistically can go together.

>The AI doesn't "know" what it "knows". Yes it does. https://news.ycombinator.com/item?id=37874556

OK, how about we put it this way?

Who cares if "it knows" if it's apparently impossible to get it to use that knowledge to stop hallucinating?

To me, the end user there is no practical difference between not having the data and not being able to use the data it has. If it can't use it, or refuses to use it, it may as well not exist.

Re: "Hallucinating" AIs sound creative, but let's not celebrate being wrong

#98

Post by teaearlgraycold »

pdonis wrote: post #65
teaearlgraycold wrote: post #16

Earlier quoted context omitted.

> LLMs have no understanding of the underlying reality Not true. They are slowly gaining an understanding of reality by reverse engineering the relationships built into human languages. The only reason LLMs are getting better is because they are better modeling the world. At some point the only way to improve token prediction is to gain an understanding of the world.

> They are slowly gaining an understanding of reality by reverse engineering the relationships built into human languages. Perhaps the people building LLMs are doing this, but the LLMs themselves are not. LLMs are just generating text. They aren't "reverse engineering" anything. > The only reason LLMs are getting better is because they are better modeling the world. No, they are getting better at generating text that…

Here's a bit from an episode of This American Life that I think is relevant, cut into a post on my blog:

https://danangell.com/blog/posts/gpt-understands/

> The idea is that back when GPT-4 was being trained for it to really consistently get the next word correct, to do that reliably, it had to do more than just bullshit. It had to do more than guess based on patterns. To get the next word right, it had to truly understand the words coming before it.

Imagine if you were training a transformer to predict numbers in a sequence that came from a sine function. Eventually it would get a pretty good test score and you'd find that internally it had built a decent replication of the sine function.

Now do the same with English text. Throw enough data and computation at it and eventually it'll model a rough understanding of gravity and the shapes of objects, as seen in the different answers between GPT-3.5 and GPT-4.

Re: "Hallucinating" AIs sound creative, but let's not celebrate being wrong

#99

Post by ryukoposting »

Spot-on article, for the most part. I firmly believe that the use of an LLM as a fact-finding tool represents a misunderstanding of the technology's abilities and intent. If you're using an LLM like it's a search engine, then you should just use a search engine instead. LLMs are conversational devices. Search engines are an index of human knowledge. Right tool, right job.

I'm fine with the term "hallucination," though. Hallucinations are, by definition, not real. The term emphasizes a detachment from reality that LLMs possess, and the general public is just beginning to grasp.

Re: "Hallucinating" AIs sound creative, but let's not celebrate being wrong

#100

Post by pnpnp »

ryukoposting wrote: post #99

Spot-on article, for the most part. I firmly believe that the use of an LLM as a fact-finding tool represents a misunderstanding of the technology's abilities and intent. If you're using an LLM like it's a search engine, then you should just use a search engine instead. LLMs are conversational devices. Search engines are an index of human knowledge. Right tool, right job. I'm fine with the term "hallucination," thoug…

I’m not sure I agree with this one. My Google results have gotten pretty awful lately.

I’ll usually go to an LLM for a very specific search term, then Google to find a reference or validate.

Post reply on HN