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

Google scrambles to manually remove weird AI answers in search

theverge.com

Post reply on HN
Print view View original

151–160 of 387 posts

Re: Google scrambles to manually remove weird AI answers in search

#151

Post by willis936 »

jerf wrote: post #27

"Achieving the initial 80 percent is relatively straightforward since it involves approximating a large amount of human data, Marcus said, but the final 20 percent is extremely challenging. In fact, Marcus thinks that last 20 percent might be the hardest thing of all." 100% completely accurate is super-AI-complete. No human can meet that goal either. No, not even you, dear person reading this. You are wrong about som…

Failing loudly is an excellent feature. "More compelling lies" is not the answer.

Re: Google scrambles to manually remove weird AI answers in search

#152

Post by madeofpalk »

lukev wrote: post #115
jononor wrote: post #107

Earlier quoted context omitted.

Encyclopedia Britannica is also wrong in a reproducible and fixable way. And the input queries a finite set. It's output does not change due to random or arbitrary things. It is actually possible to verify. LLMs so far seem to be entirely unverifiable.

They don’t just seem it. They are by design. We talk about models “hallucinating” but that’s us bringing an external value judgement after the fact. The actual process of token generation works precisely the same. It’d be more accurate to say that models always hallucinate.

Yes - this is what i've been saying all the time. The term 'hallucinations' is misleading because the whole point of LLMs is that they recombine all their inputs into something 'new'. They only ever hallucinate outputs - that's their whole point!

Re: Google scrambles to manually remove weird AI answers in search

#154

Post by threeseed »

kenjackson wrote: post #146
threeseed wrote: post #132

Earlier quoted context omitted.

I can ask a human to explain the steps they took to answer a question. I can ask a human a question 100 times and I don't get back 100 different answers. None of those applies to an LLM.

Ask a human what the meaning of life is and how it impacts their day to day interactions. I know I can tell you an answer but I couldn’t tell you steps about how I got it. And if you asked it to me twice I’d definitely give different answers unless you told me to give the same answer. In part I’d give a different answer because if someone asks me the same question twice I assume the first answer wasn’t sufficient.

No one is taking about existential questions about meaning of life.

We are talking about basic things like whether or not to eat rocks or put glue in recipes. We can answer those questions with a chain of logic and repeatability.

Re: Google scrambles to manually remove weird AI answers in search

#155

Post by soloist11 »

ta8645 wrote: post #144
threeseed wrote: post #132

Earlier quoted context omitted.

I can ask a human to explain the steps they took to answer a question. I can ask a human a question 100 times and I don't get back 100 different answers. None of those applies to an LLM.

You can ask an LLM to explain itself.. it will give you a logical stepwise progression from your question to its answer. It will often contain a mistake, but the same is true for a human. And if your LLM is giving you 100 different answers, then it has been configured to do so. Because instead, it could be configured to never vary at all. It could be 100% reproducible if so desired.

[dead]

Re: Google scrambles to manually remove weird AI answers in search

#156

Post by ImAnAmateur »

39896880 wrote: post #147
zihotki wrote: post #78

Earlier quoted context omitted.

They spent 10 years finetuning the search and then another 15 finetuning ads and clicks. Google's business is ads, not search.

Apologies in advance for this level pedantry: Google’s business is behavioral futures, not ads. Ads are just a means to that particular end.

Surveillance capitalism? What are behavioral futures?

Re: Google scrambles to manually remove weird AI answers in search

#157

Post by empath75 »

I love chatgpt and use it all the time and find it tremendously useful, but I never want to see AI generated content when I am not specifically looking for it. I don't want to see it in comments, I don't want to see it in search results, I don't want to see it as an illustration for an article, I _really_ don't want to see AI generated word vomit blog posts or fake "news" articles when I'm looking for actual information.

It's not even because it's sometimes (or often) wrong or full of hallucinations. Even if it's 100% factually correct all of the time, it's _poor quality writing and art_, full of cliches and bland generalities, which even if they solve all the rest of the problems it's sort of fundamental to the architecture of transformers. You can't ever be truly creative or unique if you're predicting the _most likely_ token.

Re: Google scrambles to manually remove weird AI answers in search

#158

Post by reustle »

rurp wrote: post #150
PreInternet01 wrote: post #63

It's debatable whether Google has truly lost the plot because of the "AI wars", but the moment the statement "Bing returns more sensible results than you" becomes verifiably true, it's... cause for concern? The approach that Google appears to have taken, which is to assume that the top-ranked part of its current search index is a sensible knowledge base, may have been true some years ago, but definitely isn't now: fo…

Not only is the current internet 80% spam, it's rapidly approaching 99% thanks in large part to LLMs. At this point I would be shocked if Google had a solid plan for how to handle this going forward as the problem space gets more difficult.

https://en.wikipedia.org/wiki/Dead_Internet_theory

Re: Google scrambles to manually remove weird AI answers in search

#159

Post by dleeftink »

somenameforme wrote: post #123
jononor wrote: post #107

Earlier quoted context omitted.

Encyclopedia Britannica is also wrong in a reproducible and fixable way. And the input queries a finite set. It's output does not change due to random or arbitrary things. It is actually possible to verify. LLMs so far seem to be entirely unverifiable.

LLMs are completely deterministic even if that's kind of weird to state because they output things in terms of probabilities. But if you simply took the highest probability next word, you'd always yield the exact same output given the exact same input. Randomness is intentionally injected to make them seem less robotic through the 'temperature' parameter. Why it's not just called the rng factor is beyond me.

Maybe some models can be deterministic at a point in time, but train it for another epoch with slight parameter changes and a revised corpus and determinism goes out the proverbial (sliding) window real quick. This is not unwanted per se, and the exact feedback loop that needs improving to better integrate new knowledge or revise knowledge artefacts incrementally/post-hoc.

Re: Google scrambles to manually remove weird AI answers in search

#160

Post by maximinus_thrax »

threeseed wrote: post #143
SoftTalker wrote: post #5

Trained on Twitter and Reddit. Garbage in/Garbage out, as it has always been.

Except that 90% of Reddit isn't garbage. It's really useful. Problem is Google can't tell what is garbage or not. No LLM can.

> Except that 90% of Reddit isn't garbage. It's really useful.

Citation needed. I've been a Reddit user since its inception and honestly except for niche hobby subreddits, Reddit is mostly low effort garbage, bots and rehashed content. I'd wager that mainstream subreddits are 99% garbage for training an LLM for anything other than shitposting.

Post reply on HN