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

ChatGPT Search

openai.com

Post reply on HN
Print view View original

991–1000 of 1001 posts

Re: ChatGPT Search

#991

Post by marcus_holmes »

belter wrote: post #836
marcus_holmes wrote: post #651

Earlier quoted context omitted.

I agree that this is the core question, but I'd put it as: Who Gets To Decide What Is True? With a search paradigm this wasn't an issue as much, because the answers were presented as "here's a bunch of websites that appear to deal with the question you asked". It was then up to the reader to decide which of those sites they wanted to visit, and therefore which viewpoints they got to see. With an LLM answering the que…

> Who Gets To Decide What Is True? Seems particularly to be a US based phenomenon. Unlike the more transparent manipulation seen under dictatorships... Where people generally recognize propaganda for what it is, even if they can’t openly challenge it—some in USA live within entirely new realities.

I'm not in the USA and I see this a lot too. It's not just a US-based thing.

Re: ChatGPT Search

#992

Post by webninja »

marcus_holmes wrote: post #651
freediver wrote: post #241

Been thinking about this a lot [1]. Will this fundamentally change how people find and access information? How do you create an experience so compelling that it replaces the current paradigm? The future promised in Star Trek and even Apple's Knowledge Navigator [2] from 1987 still feels distant. In those visions, users simply asked questions and received reliable answers - nobody had to fact-check the answers ever. C…

I agree that this is the core question, but I'd put it as: Who Gets To Decide What Is True? With a search paradigm this wasn't an issue as much, because the answers were presented as "here's a bunch of websites that appear to deal with the question you asked". It was then up to the reader to decide which of those sites they wanted to visit, and therefore which viewpoints they got to see. With an LLM answering the que…

> To paraphrase a recent conversation I had with a friend: "in the USA, can illegal immigrants vote?" has a single truthful answer ("no" obviously). But there are many places around the web saying other things (which is why my friend was confused).

This doesn’t have a single truthful answer. Some states don’t have voter ID laws, so the truth can depend on the state. In those no voter ID laws there’s not that much keeping someone from voting twice or more under different names, except significant moral qualms about subverting instead of preserving everyone else’s right to vote in a democratic republic. Someone can assume the name of a person from another country that could’ve plausibly come in illegally. Without a picture ID, they can’t claim you aren’t that persons.

Can an illegal immigrant vote? Yes, in states without voter ID laws, technically anyone can vote, even convicted terrorists. Should an illegal immigrant vote? No, they’re not supposed to be able to vote and there may be consequences if caught.

What purpose do a lack of voter ID laws serve except the obvious conclusion which is to enable cheating?

Re: ChatGPT Search

#993

Post by seanmcdirmid »

webninja wrote: post #992
marcus_holmes wrote: post #651

Earlier quoted context omitted.

I agree that this is the core question, but I'd put it as: Who Gets To Decide What Is True? With a search paradigm this wasn't an issue as much, because the answers were presented as "here's a bunch of websites that appear to deal with the question you asked". It was then up to the reader to decide which of those sites they wanted to visit, and therefore which viewpoints they got to see. With an LLM answering the que…

> To paraphrase a recent conversation I had with a friend: "in the USA, can illegal immigrants vote?" has a single truthful answer ("no" obviously). But there are many places around the web saying other things (which is why my friend was confused). This doesn’t have a single truthful answer. Some states don’t have voter ID laws, so the truth can depend on the state. In those no voter ID laws there’s not that much kee…

That isn’t true. They do have voter registration, and they match name addresses that are cross them off, a name can’t vote twice.

MAGA types who wanted to prove democrats could cheat like this were caught very quickly because voting twice or voting for someone else is very very easy to detect, even without IDs. The judges gave them Darwin awards.

Unregistered voters can’t vote in any state, they can only do so by pretending to be someone else (name and address matches the rolls), but they are caught when those people actually vote. Foreigners can also vote by illegally registering. But motor voter means they actually check your status at registration.

The ID thing is a solution to a non-problem, like literacy tests were. But the republicans could make it easier to achieve universally if they just went with a guaranteed free national ID like other countries, except that would make the obstruction aspect of voter ID requirements much more moot, so they never go there.

Foreign residents can vote in some local election in a few places, mainly very local school board elections.

Re: ChatGPT Search

#994

Post by quantamiser »

most times I dont use search to actually do "research". I just put a company's or location's name because I dont remember their website and open the top non-ad result. ChatGPT's way of publishing the result might be friction in this case

Re: ChatGPT Search

#995

Post by afh1 »

marcus_holmes wrote: post #651
freediver wrote: post #241

Been thinking about this a lot [1]. Will this fundamentally change how people find and access information? How do you create an experience so compelling that it replaces the current paradigm? The future promised in Star Trek and even Apple's Knowledge Navigator [2] from 1987 still feels distant. In those visions, users simply asked questions and received reliable answers - nobody had to fact-check the answers ever. C…

I agree that this is the core question, but I'd put it as: Who Gets To Decide What Is True? With a search paradigm this wasn't an issue as much, because the answers were presented as "here's a bunch of websites that appear to deal with the question you asked". It was then up to the reader to decide which of those sites they wanted to visit, and therefore which viewpoints they got to see. With an LLM answering the que…

Unfortunately most people seem comfortable with the idea of a Ministry of Truth, at least here in Brazil and from what I read on the web, in many parts of the world also.

Re: ChatGPT Search

#997

Post by mvkel »

umvi wrote: post #451
whizzter wrote: post #300

This is probably Google's Altavista moment, by making their results crappier by the year in search of Ad dollars everyone has felt that there is a potential for search to be better and once that becomes available they'll be in a continious game of catch-up. Yes, Google has their own AI divisions, tons of money and SEO is to blame for part of their crappiness. But they've also _explicitly_ focused on ad-dollars over a…

At some point it seems like Google switch to ML-based search instead of index based search. You can search for very specific combinations of lyrics and scenes: "eyes on me pineapple bucket of water house of cards chess time loop" and you won't surface a link to the music video featuring all of those things ( https://www.youtube.com/watch?v=AlzgDVLtU6g ), you'll just get really generic results of the average of your q…

The question of how to improve the google experience for users has been clear for years, possibly a decade.

The question that remains unanswered is how google can do so without compromising customer revenue.

Re: ChatGPT Search

#998

Post by mmahemoff »

rany_ wrote: post #978
DrammBA wrote: post #958

Earlier quoted context omitted.

Why is it that comments complaining about flaky AI responses never share the chat link, and then invariably someone replies with examples of the AI answering correctly but also fail to share the chat link?

I just forgot that this feature exists. At any rate, isn't it more convenient to share the output directly? What's the difference? 4o-mini: https://chatgpt.com/share/6725f6fb-3b84-800b-9b4d-3e82554fbf... I can't figure out how to share the 4o one. The UI is weird.

404 Not Found

Re: ChatGPT Search

#999

Post by davideg »

Slyfox33 wrote: post #987
davideg wrote: post #678

Earlier quoted context omitted.

Several folks have already mentioned that ChatGPT with search returns the correct answer to your question (with a source to explore directly). I really think this latest release is a game changer for ChatGPT since it seems much more likely to return genuine information than ChatGPT answering using its model alone. Of course it still hallucinates sometimes (I asked about searching tabs in Firefox Mobile and it told hi…

Chat gpt will not always return the correct answers, thats a fundamental limitation of how it works since its non deterministic. So just saying "it worked for me" means nothing.

Yes and having direct links to sources to validate its answers is a huge benefit given the fundamental limitations of LLMs. Now you can decide if you trust the source material directly.

Re: ChatGPT Search

#1000

Post by fennecbutt »

freediver wrote: post #241

Been thinking about this a lot [1]. Will this fundamentally change how people find and access information? How do you create an experience so compelling that it replaces the current paradigm? The future promised in Star Trek and even Apple's Knowledge Navigator [2] from 1987 still feels distant. In those visions, users simply asked questions and received reliable answers - nobody had to fact-check the answers ever. C…

I do love how all the comments on the top comment of every HN post just consists of deeper and deeper levels of inane bickering ahaha. It's like a template.
Post reply on HN