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

171–180 of 454 posts

Re: Kagi News

#171

Post by nicbou »

whatamidoingyo wrote: post #123
__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…

> 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.

I monitor 404 errors on my website. ChatGPT frequently sends traffic to pages that never existed. Sometimes the information they refer to has never existed on my website.

For example: "/glossary/love-parade" - There is no mention of this on my website. "/guides/blue-card-germany" has always been at "/guides/blue-card". I don't know what "/guides/cost-of-beer-distribution" even refers to.

Re: Kagi News

#173

Post by jrowen »

This doesn’t really seem to touch on the problem I have with news, which is that it is all doom and gloom, FUD and outrage. The headlines I saw:

Trump, Congress deadlock as shutdown deadline nears

Taliban cuts internet nationwide, flights grounded in Afghanistan

Indonesia school collapse leaves 38 missing, 77 hurt

YouTube settles Trump suspension lawsuit for $24.5m

German court jails AfD aide for China spying

US deports 120 Iranians after deal

Russian drone strike kills family of four

Is this really what I need to know in the world? Am I saying “informed”? This is not helping the anxiety from reading news described in the article. This is not good for people.

Re: Kagi News

#174

Post by rsingel »

e40 wrote: post #42

I'm not affiliated, but I've used https://www.memeorandum.com/ for more than a decade. Gives me a good high-level view of the news. I'm a Kagi customer and I definitely don't want anything they do with the news.

Gabe Rivera has been doing that right for almost 20 years.

If you haven't seen it there's also an amazing feature that you can go back and see the homepage as it was from any point in time in the last 20 years

Re: Kagi News

#175

Post by graynk »

rsingel wrote: post #165

Looks like yet another aggregator that will send almost no referral traffic to the folks who actually put labor into writing. Parasitic by definition. And embracing the news from nowhere perspective. So both a parasite and boring at the same time. I wish more tech folks who want to "fix the news" would learn from Gabe Rivera's Techmeme, Memeorandum, and Mediagazer. He's done aggregation right for 20 years

It uses RSS from the folks who actually put labor into writing and links to the sources under every article.

Re: Kagi News

#176

Post by kevcampb »

> 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.

Re: Kagi News

#177

Post by vbezhenar »

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.

They could make up source, but ChatGPT is an actual app with complicated backend, not dumb pipe between textedit and GPU. Surely they could verify on server side every link they output to user before including it in the answer. I'm sure Codex will implement it in no time!

Re: Kagi News

#178

Post by b112 »

arrosenberg wrote: post #155
NobodyNada wrote: post #124

Earlier quoted context omitted.

I'm about as AI-pessimist as it gets, but Kagi's use of LLMs is the most tasteful and practical I've seen. It's always completely opt-in (e.g. "append a ? to your search query if you want an AI summary", as opposed to Google's "append a swear word to your search query if you don't want one"), it's not pushy, and it's focused on summarizing and aggregating content rather than trying to make it up.

FYI, you can append &udm=14 to Google searches to remove AI results and a bunch of the other clutter they've added.

I did that, and started getting flagged as a bot. Had to search elsewhere(Kagi) full time, or else suffer endless "find a bike" nonsense.

I think Google hates the loss of no/few ads or lame suggestions.

Re: Kagi News

#179

Post by imiric »

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…

> We need better journalism, not better summarizing aggregators.

I agree, but how do you envision that happening? Journalism died a long time ago, arguably around the birth of the 24-hour news cycle, and it was further buried by social media. A niche tech company can only provide a better way to consume what's out there, not solve such large societal problems.

> There are no "lifehacks" to getting informed.

I don't think their intent is to change how people are informed. What this aims to do is replace endless doomscrolling on sites that are incentivized to rob us of our attention and data, with spending a few minutes a day to get a sense of general events around the world. If something piques your interest, you can visit the linked sources, or research the event elsewhere. But as a way of getting a quick general overview of what's going on, I think it's great.

Re: Kagi News

#180

Post by m4r71n »

j2kun wrote: post #146
__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…

This seems like the opposite of "privacy by design" > Privacy by design: Your reading habits belong to you. We don’t track, profile, or monetize your attention. You remain the customer and not the product. But the person running the LLM surely does.

How would the LLM provider get any information about your reading habits from the app? The LLM is used _before_ the news content is served to you, the reader.
Post reply on HN