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

Kagi News

blog.kagi.com

Post reply on HN
Print view View original

241–250 of 454 posts

Re: Kagi News

#241

Post by mac-attack »

carlosjobim wrote: post #203
fresh_broccoli wrote: post #159

Earlier quoted context omitted.

Also, they talk about "echo chambers" and "full spectrum of global perspectives". Representing all perspectives sounds great in theory, but how far should it go? Should all politicians' remarks be reproduced verbatim with absolutely no commentary, no fact-checking and no context? Should an article about an airplane crossing the Pacific include "some experts believe that this is impossible because Earth is flat?" Exce…

> Should all politicians' remarks be reproduced verbatim with absolutely no commentary, no fact-checking and no context? Yes. That's an interview, and is much better than summarizations and short soundbites and one-sentence quotes.

This presupposes that you have an informed population that already knows the facts and context

Re: Kagi News

#242

Post by 4dm1r4lg3n3r4l »

Vexowsky wrote: post #233

I liked Kagi, was paying for it for a few months, but $10 is just too much

I don't find it too much. For $10 I get a search engine better than all the others, I get access to many AI models via Kagi Assistant and Kagi Translate.

While I understand different people find value in different things, dismissing Kagi generally as "too expensive" is ignorant IMO.

I think it also depends what you use it for. I use both their search and their AI models for software development and it saves me precious time when looking for information - in a way it pays for itself.

Re: Kagi News

#243

Post by timeon »

BeetleB wrote: post #51
__jonas wrote: post #23

Just to be clear I'm understanding correctly: This is pulling the content of the RSS feeds of several news sites into the context window of an LLM and then asking it to summarize news items into articles and fill in the blanks? I'm asking because that is what it looks like, but AI / LLMs are not specifically mentioned in this blog post, they just say news are 'generated' under the 'News in your language' heading, whi…

It's not binary - it's a continuum. When you go to Google News, the way they group together stories is AI (pre-LLM technology). Kagi is merely taking it one step further. I agree with your concern. I see this as a convenient grouping, and if any interests me I can skip reading the LLM summary and just click on the sources they provide (making it similar to Google News).

> When you go to Google News

You don't and you should not use this one either.

Re: Kagi News

#244

Post by 4dm1r4lg3n3r4l »

skybrian wrote: post #92

Do the checkmarks do anything? I expected them to disappear after a reload (like hiding a post on Hacker News), but apparently that's not what they're for. "Mark as read" checks all the checkmarks, but since they're still there after a reload, I don't see the point.

It automatically marks stuff you've seen - it's just a visual cue. Similar to how search engines (like Google) show visited links with a different color.

I think keeping them on the page instead of automatically hiding them makes more sense for a product that's trying to update their news feed once per day. You feel more in control, as if it's not a stream of never-ending stories, but rather a fixed amount of stories that you can realistically power through. Seeing all items checked sort-of supports this philosophy.

Re: Kagi News

#245

Post by BeetleB »

fresh_broccoli wrote: post #138

I think this is the wrong direction. We need better journalism, not better summarizing aggregators. Summaries are no substitute for real articles, even if they're generated by hand (and these apparently are not). Summaries are bound to strip the information of context, important details and analysis. There's also no accountability for the contents. Sure, there are links to the actual articles, but let's not kid ourse…

That's precisely what Axios does, and they make money from this (and they don't list their sources). So I can see Kagi pursuing this.

FWIW, I agree with you.

I used to be a news junkie. I've always thought of writing the lessons I learned, but one of them was "If you're a casual news reader, you are likely more misinformed than the one who doesn't read any news." One either should abstain or go all in.

I guess I'd amend it to put people who only glance at headlines to be even more misinformed. It was not at all unusual for me to read articles where the content just plain disagreed with the headline!

Re: Kagi News

#246

Post by 4dm1r4lg3n3r4l »

kevcampb wrote: post #176

> This is pulling the content of the RSS feeds of several news sites into the context window of an LLM and then asking it to summarize news items into articles and fill in the blanks? This is awful. It's cutting out any money going to the news agencies that go out there and write news. If they didn't exist, Kagi wouldn't work.

> This is awful. It's cutting out any money going to the news agencies that go out there and write news. If they didn't exist, Kagi wouldn't work.

Why would Kagi stop working if news didn't exist? Kagi is a search engine first and foremost, Kagi News is not a money making product of theirs. Kagi would still be making money with their search engine.

Also, this should entice news writers to write better news. The main reason people use products such as this is that they are sick and tired of going to news sites only to have to power through filler material to get the 10% that actually matters...

Re: Kagi News

#247

Post by motoxpro »

stavros wrote: post #229
edaemon wrote: post #220

Earlier quoted context omitted.

I'm not the person you asked, but it's useful to know if the summary was generated using a method prone to inaccuracy.

That's all methods, though. Have you seen humans?

In this situation, humans are more accurate, for now, so it's good information to have.

Same as I would like to know if humans self assessed in a study about how well they drive vs the empirical evidence. Humans just aren't that good at that task so it would be good to know coming in.

Just call it Kagi Vibes instead of Kagi News as news has a higher bar (at least for me)

Re: Kagi News

#249

Post by BeetleB »

Vexowsky wrote: post #233

I liked Kagi, was paying for it for a few months, but $10 is just too much

I get by with the $5 plan. I don't use it at work, and it shows. I often curse during work that I can't exclude certain domains permanently from my Google search!

Re: Kagi News

#250

Post by Paradigma11 »

CaptainOfCoit wrote: post #152
whatamidoingyo wrote: post #123

Earlier quoted context omitted.

> when you ask an LLM to point to 'sources' for the information it outputs, as far as I know there is no guarantee that those are correct A lot of times when I ask for a source, I get broken links. I'm not sure if the links existed at one point, or if the LLM is just hallucinating where it thinks a link should exist. CDN libraries, for example. Or sources to specific laws.

> A lot of times when I ask for a source, They'll do pretty much everything you ask of them, so unless the text actually come from some source (via tool calls, injecting content into the context or other way), they'll make up a source rather than doing nothing, unless prompted otherwise.

Wrong, just ask it about some non existent famous historical person and it will most likely tell you it didnt exist.
Post reply on HN