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

271–280 of 1001 posts

Re: ChatGPT Search

#271

Post by PittleyDunkin »

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…

> In those visions, users simply asked questions and received reliable answers - nobody had to fact-check the answers ever.

This also seems like a little ridiculous premise. Any confident statement about the real world is never fully reliable. If star trek were realistic the computer would have been wrong once in a while (preferably with dramatically disastrous consequences)—just as the humans it likely was built around are frequently wrong, even via consensus.

Re: ChatGPT Search

#272

Post by lawn »

randcraw wrote: post #87

I really don't see the value of summarizing/repackaging web search hits. Given that 99% of SEO-tuned web content is just shilling for vendors who don't want to be seen, LLM search summarization will just repackage those ads into a more palatable format that is LESS useful than the original, while more successfully hiding the obvious signatures that used to be a clear warning sign that... THE.FOLLOWING.CONTENT.IS.MANI…

To me the kombination of Kagi's good search results with their AI summarizer has been very useful.

Of course, layering an LLM on top of garbage will still produce garbage.

Re: ChatGPT Search

#273

Post by kevin_thibedeau »

fulafel wrote: post #198
cjf101 wrote: post #183

If the current iteration of search engines are producing garbage results (due to an influx of garbage + SEO gaming their ranking systems) and LLMs are producing inaccurate results without any clear method proposed to correct them, why would combining the two systems not also produce garbage? The problem I see with search is that the input is deeply hostile to what the consumers of search want. If the LLM's are partic…

Garbage-ness of search results is not binary, the right question is: can LLMs improve the quality of search results? But sure, it won't end the cat and mouse game.

> it won't end the cat and mouse game.

There is no way to SEO the entire corpus of human knowledge. ChatGPT is very good for gleaning facts that are hard to surface in today's garbage search engines.

Re: ChatGPT Search

#274

Post by ColinHayhurst »

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…

As Porter explained in 1980 there are three ways to compete successfully:

1. On price; race to the bottom or do free with ads

2. Differentiation

3. Focus - targeting a specific market segment

Some things don't change. Land grabbers tend to head down route 1.

Re: ChatGPT Search

#275

Post by numbers »

ryzvonusef wrote: post #217

https://x.com/sahir2k/status/1852038475606036577 > how tf is it reading private repos ?!

Not sure how it was able to read this user's repo name but for me, it's optimistically saying "yes they have a repo named X" of whatever I ask it and sometimes I do have that repo and sometimes I don't.

I have a private repo named "portland-things" and I asked "does this user have a repo related to portland?" and it responded with "yes it's called 'pdx'" but that's not correct at all.

Re: ChatGPT Search

#276

Post by valval »

cjf101 wrote: post #183

If the current iteration of search engines are producing garbage results (due to an influx of garbage + SEO gaming their ranking systems) and LLMs are producing inaccurate results without any clear method proposed to correct them, why would combining the two systems not also produce garbage? The problem I see with search is that the input is deeply hostile to what the consumers of search want. If the LLM's are partic…

I’d be more cynical still and ask, where is correct information found in the first place? Humans of all shape and size have biases. Most research is faulty, fabricated, or not reproducible. Missing information tells a greater story than existing one.

We don’t have a way of finding objective information, why would we be able to train a model to do so?

Re: ChatGPT Search

#277

Post by azinman2 »

jtgverde wrote: post #253
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…

Great find on the knowledge navigator, I had never seen it but I was a toddler when it was released haha. It's interesting how prescient it was, but I'm more struck wondering--would anyone in 1987 have predicted it would take 40+ years to achieve this? Obviously this was speculative at the time but I know history is rife with examples of AI experts since the 60s proclaiming AGI was only a few years away Is this time…

You don’t need AGI to build that experience.

I’m the past there was a lot of overconfidence in the ability for things to scale. See Cyc (https://en.m.wikipedia.org/wiki/Cyc)

Re: ChatGPT Search

#280

Post by _sys49152 »

is this going to give me lyrics or recipes link or will these categories continue to be nerfed by openai? it appears to write over its response of providing me a website link and gives me an unclickable blue name of website instead.
Post reply on HN