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

41–50 of 196 posts

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

#41

Post by wildrhythms »

fragmede wrote: post #27

Curiously enough, I don't get the same result as the author on plain ChatGPT-4. https://chat.openai.com/share/e213e0bd-2838-45e2-9942-e52954... https://chat.openai.com/share/64bc62d3-042c-40c4-8d50-8e28ce... Hilariously, plugging the example in the article into Bing enhanced ChatGPT-4, ChatGPT-4 w/ Bing hallucinates because of that very article! https://chat.openai.com/share/4fe50933-8436-44ad-a778-6297ca... If you t…

Isn't the whole point of these LLMs be that they are non-deterministic? And with OpenAI opaquely 'tweaking' the model, it's no surprise you don't get the same output.

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

#43

Post by Barrin92 »

it's just a consequence of the annoying anthropomorphizing of tech. "Human's can't do X, the AI model can't do X, look it's just like me fr, fr". Of course nobody applies that logic to a forklift or a debugger. If gdb started to hallucinate variables into existence we don't call it a creative act, we call it a bug.

Given that these AI systems just like any other machine operate at scale, automated, and fast, they must be precise and transparent, that is where the work should be.

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

#44

Post by jshen »

happytiger wrote: post #11

The LLM AI technology generation is optimized to be fluently conversational and not to be factually correct all the time. 1) Hallucinations often appear because LLMs are designed to create fluent, coherent text. 2) LLMs have no understanding of the underlying reality that language describes. 3) LLMs use statistics to generate language that is grammatically and semantically correct within the context of the prompt. It…

> "The LLM AI technology generation is optimized to be fluently conversational and not to be factually correct all the time."

I always find this point a bit odd because humans aren't "optimized" to be correct either.

> "LLMs have no understanding of the underlying reality"

I struggle with this one because I see both sides of it. I was making a prompt the other day and gave a CSV file as an input and told the LLM if could only answer with values from one column and it did exactly as I asked. It's hard for me to see things like that and not believe it has an understanding at some level.

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

#45

Post by gojomo »

Retric wrote: post #39
gojomo wrote: post #32

Contra the headling, we should "celebrate being wrong" – when it's wrong in fast, interesting, & voluminous ways that can then be filtered or corrected. That's how lots of science, innovation, & learning work: generate many superficially-plausible candidates via a fast-and-loose process, then refine with a more rigorous evaluation. That AIs, in the form of LLMs, are now doing this so well was unexpected, and progress…

Science progresses because people are testing plausible ideas not random ideas. We don’t do human trials on random drugs, we do human trials on promising ones. Animal trials are more open but even then candidates are carefully considered as being viable. Things are even more open the earlier you are in drug discovery, but we aren’t testing molecules using Dubnium or most elements on the periodic table. Similarly Ecol…

Yes - but even bulk LLM 'hallucinations' are usually plausible, and far from 'random'.

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

#46

Post by Jensson »

fragmede wrote: post #27

Curiously enough, I don't get the same result as the author on plain ChatGPT-4. https://chat.openai.com/share/e213e0bd-2838-45e2-9942-e52954... https://chat.openai.com/share/64bc62d3-042c-40c4-8d50-8e28ce... Hilariously, plugging the example in the article into Bing enhanced ChatGPT-4, ChatGPT-4 w/ Bing hallucinates because of that very article! https://chat.openai.com/share/4fe50933-8436-44ad-a778-6297ca... If you t…

> This conversation may reflect the link creator’s personalized data, which isn’t shared and can meaningfully change how the model responds.

You have side channel data, so you didn't run the same thing he did.

If you want to run the same test then you have to clear your personalized data as well.

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

#47

Post by hawski »

When I hear the word "hallucination" in my mind appears an image of a crazy guy almost with foam on his mouth, probably on drugs or having severe mental problems. That is not a thing that I associate with being creative and certainly not trust worthy.

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

#48

Post by delecti »

BryanLegend wrote: post #38

All thoughts are hallucinations. https://en.wikipedia.org/wiki/Cogito,_ergo_sum

"I think, therefore I am" means that one of the only things you can be sure of is that, as a thinking entity, you must exist. Thinking the thought "I am thinking", requires that you exist to do that thinking.

So unless someone was arguing that LLMs didn't exist, that principle is pretty orthogonal to the discussion.

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

#49

Post by 98codes »

happytiger wrote: post #11

The LLM AI technology generation is optimized to be fluently conversational and not to be factually correct all the time. 1) Hallucinations often appear because LLMs are designed to create fluent, coherent text. 2) LLMs have no understanding of the underlying reality that language describes. 3) LLMs use statistics to generate language that is grammatically and semantically correct within the context of the prompt. It…

When a human reads a bunch of stuff on the internet and regurgitates it as fact, we don't say they hallucinate, we say that they're wrong.

Why we baby AI on this front, I have no idea.

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

#50

Post by Retric »

gojomo wrote: post #45
Retric wrote: post #39

Earlier quoted context omitted.

Science progresses because people are testing plausible ideas not random ideas. We don’t do human trials on random drugs, we do human trials on promising ones. Animal trials are more open but even then candidates are carefully considered as being viable. Things are even more open the earlier you are in drug discovery, but we aren’t testing molecules using Dubnium or most elements on the periodic table. Similarly Ecol…

Yes - but even bulk LLM 'hallucinations' are usually plausible, and far from 'random'.

What seems plausible depends on your level of suspect matter expertise.

Hallucinations are in general ridiculously incorrect and nowhere close to anything worth testing. Put another way what percentage of molecules are worth testing as a viable treatment for epilepsy? 1 in 100 trillion, less? Do you really expect hallucinations to generally pick both plausible and untested targets here?

Post reply on HN