Kagi News

Post by sally_glance » Tue, Sep 30, 2025, 9:23 PM UTC

freediver wrote: post #282

mvieira38 wrote: post #47

Earlier quoted context omitted.

Yes, that's what it is. Kagi as a brand is LLM-optimist, so you may be fundamentally at odds with them here... If it lessens the issue for you, the sources of each item are cited properly in every example I tried, so maybe you could treat it as a fancy link aggregator

> Kagi as a brand is LLM-optimist Kagi founder here. I am personally not an LLM-optimist. The thing is that I do not think LLMs will bring us to "Star Trek" level of useful computers (which I see humans eventually getting to) due to LLM's fundamentally broken auto-regressive nature. A different approach will be needed. Slight nuance but an important one. Kagi as a brand is building tools in service of its users, no p…

Post by alexsmirnov » Tue, Sep 30, 2025, 9:28 PM UTC

  Miniflux (https://miniflux.app/) in Docker, fetching 75 RSS feeds I've collected over the years

  ~200 lines in a Jupyter notebook:
  - Fetch entries from Miniflux API (last 24-48 hours)
  - Convert to CSV, feed to LLM. GPT-5 identifies trending stories across sources
  - Each article gets web-fetched and summarized via Gemini-2.5-flash
  - Results render via IPython.display

  Ten minutes per day, fully informed.

Post by tantalor » Tue, Sep 30, 2025, 9:29 PM UTC

giorgio-kagi wrote: post #309

tantalor wrote: post #79

The grounding is really strange. This example includes a Reddit post as a source: https://kite.kagi.com/s/hjgy55 But that post is actually a link to reuters.com There is also a list of "citations" which are referenced from the generated text, and "sources" which are not referenced anywhere. It's not clear if they used reddit or reuters to generate any of the text. I also see lots of citations to "common knowledge"...…

Hey - Reddit is included as a source because posts can contain firsthand accounts and independent analysis. The RSS feed we use from Reddit provides the post title and text. So to answer your question, when you see Reddit as a source we are using the aforementioned data, not what it links to. About "common knowledge" sources - we validate all content for accuracy. When the LLM needs to add context that's missing from…

Post by malfist » Tue, Sep 30, 2025, 9:31 PM UTC

MarcelOlsz wrote: post #308

Harmon758 wrote: post #211

Earlier quoted context omitted.

Just for concrete confirmation that LLM(s) are being used, there's an open issue on the GitHub repository, on hallucinations with made up information, where a Kagi employee specifically mentions "an LLM hallucination problem": https://github.com/kagisearch/kite-public/issues/97 There's also a line at the bottom of the about page at https://kite.kagi.com/about that says "Summaries may contain errors. Please verify imp…

Man am I tired of this stuff.

Post by fgalhwrg » Tue, Sep 30, 2025, 9:39 PM UTC

andrewinardeer wrote: post #301

byearthithatius wrote: post #269

Earlier quoted context omitted.

It is getting easier and easier to fake stuff and there are becoming less and less fully trusted institutions. So sadly I think you are right. Its scary but we are likely heading towards a future where you need to pay to get verified information and that itself will likely be segmented to different subscriptions for what information you want.

> It is getting easier and easier to fake stuff This is why the moon landing hoax was revolutionary in the 60's. The size of this project was enormous.