Mini feedback - it appears to report google news results as if from google and not the website in question (Wired in my case, the snapdragon x2 elite article).
Apart from that, it's really nice! Good job, kagi team!
Kagi News
21–30 of 454 posts
Re: Kagi News
#22Very skeptical that this would work for me. None of the topics that Kagi chooses to "cover" in their seven or so stories for the day resonates with what I'd want to read. That's exactly why we have feeds that you can tune to your tastes and so on. Getting rid of endless scrolling and such might be a good thing though.
Re: Kagi News
#23Just 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, which seems to imply that is what they are doing.
I'm a little skeptical towards the approach, 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 – and it does seem like sometimes they just use pure LLM output, as no sources are cited, or it's quoted as 'common knowledge'.
Re: Kagi News
#24> One daily update: We publish once per day around noon UTC, creating a natural endpoint to news consumption. This is a deliberate design choice that turns news from an endless habit into a contained ritual. I might not agree with all decisions Kagi makes, but this is gold. Endless scrolling is a big indicator that you're a consumer not a customer.
> Endless scrolling is a big indicator that you're a consumer not a customer.
Someone recently highlighted the shift from social networks to social media in a way I'd never thought about:
>> The shift from social networks to social media was subtle, and insidious. Social networks, systems where you talk to your friends, are okay (probably). Social media, where you consume content selected by an algorithm, is not. (immibis https://news.ycombinator.com/item?id=45403867)
Specifically, in the same way that insufficient supply of mortgage securities (there's a finite number of mortgages) led to synthetic CDOs [0] in order to artificially boost supply of something there was a market for.
Social media and 24/7 news (read: shoving content from strangers into your eyeballs) are the synthetic CDOs of content, with about the same underlying utility.
There is in fact a finite amount of individually useful content per unit of time.
[0] If you want the Michael Lewis-esque primer on CDOs https://m.youtube.com/watch?v=A25EUhZGBws
Re: Kagi News
#25I would love to replace google/apple news, but publishing once daily doesn't work for me.
Re: Kagi News
#26Ground offers a paid service that summarizes a broad range of sources and intentionally helps you escape your filter bubble, https://ground.news/.
Re: Kagi News
#27RSS is a strange choice in 2025. As a search engine they are in the position to extract things from web pages themselves. They already need this capability in order to properly rank the page.
For one, I welcome RSS (back) and say the more, the better. I much prefer to pull specific types of information at my own set intervals when I need them, instead of either having undifferentiated information pushed on me continuously like a blast from a fire hose, or having to reach out to manually check and filter many individual sources. The idea is to schedule my receipt and processing of the information, and then refine the stream itself as well as the intervals I use to view it and the total amount of time I spend on it.
I'm currently on the hunt for an RSS reader that has good filtering and sorting functionality, so I can (for instance) pull several feeds from only certain sources, but not see any posts/articles about terms A or B, yet see and sort any posts with term C by time, followed by either posts from source 1 with terms C and D, or posts from source 2 with terms E or F but not G, which would be sorted by relevance.
I know that's a complicated and probably poorly written explanation; but I'm imagining something like Apple Mail Rules for RSS.
Re: Kagi News
#28RSS is a strange choice in 2025. As a search engine they are in the position to extract things from web pages themselves. They already need this capability in order to properly rank the page.
Why? I do use it and can't imagine following anything without it... And I keep hoping that it will come back and replace absolutely terrible schizophrenic feeds from meta/x/etc
>Why?
Because not every site has a RSS feed. For example when Claude Sonnet 4.5 released it would make sense to have that, but there is no RSS feed for Anthropic. Being compatible with the entire web instead of just a subset of it is useful.
Re: Kagi News
#29This looks awesome, but there's absolutely no way I'm installing an app for this, sorry.
Re: Kagi News
#30Big news junkie but I don't feel the need to buy into Kagi's ecosphere personally as a SearXNG user. The article touches on signal over noise and I have found two solutions that work for me as a news junkie:
News Minimalist [1] and Boring Report [2]. Both aggregate news and (IMO) most importantly provide links from multiple outlets for the same stories. Really made me notice the clickbait and allows me to be more selective in choosing reputable sources.
Both use AI, with the former ranking news based on importance, while the latter summarizes articles. (That doesn't feel useful for supporting journalism as a whole so I typically click through and read the articles unless I don't like the outlet reporting)