Earlier quoted context omitted.
One thing I miss from Google is a reliable built-in calculator. I know I could just use some other calculator app but the habit of just typing stuff into my address bar is hard to shake. Kagi also has a calculator, but for a lot of queries it gives questionable results, for example for `210/8` it returns `105/4`. Technically correct, sure, but almost never what I want.
Just tested this out, you can add "in decimal" to get it as a decimal.
Kagi Sidekick (alpha)
141–150 of 172 posts
Re: Kagi Sidekick (alpha)
#142Earlier quoted context omitted.
Unless you design the LLM yourself and purposefully watermark the output. https://arxiv.org/pdf/2306.04634.pdf
That's a digital signature, same as sending an email with GPG to prove you sent it. You wouldn't say that because some people use GPG you can somehow detect who wrote every email on earth, it's a push model vs pull. This is why I wrote "any sentence" vs "some sentences".
I have my doubts about the effectiveness of this method and realistically, it won't make any difference because the bad actors will just use an LLM that doesn't snitch on them, so you're technically correct.
Re: Kagi Sidekick (alpha)
#143Earlier quoted context omitted.
That's a digital signature, same as sending an email with GPG to prove you sent it. You wouldn't say that because some people use GPG you can somehow detect who wrote every email on earth, it's a push model vs pull. This is why I wrote "any sentence" vs "some sentences".
Watermarking is not at all like a digital signature and a lot like steganography. I only have a surface level understanding of the process, but it works by biasing token selection to encode information into the resulting text in a way that's resistant to later modifications and rephrasing. I have my doubts about the effectiveness of this method and realistically, it won't make any difference because the bad actors wi…
Re: Kagi Sidekick (alpha)
#144Earlier quoted context omitted.
Didn't know that and thanks for giving us the idea. Aibó sounds much better than Sidekick, we may need to rename :)
Please don't write it Aibó, though. The proper romanization is Aibō or Aibou. See https://en.m.wikipedia.org/wiki/Hepburn_romanization#Long_vo...
Re: Kagi Sidekick (alpha)
#145Re: Kagi Sidekick (alpha)
#146Earlier quoted context omitted.
I suspect their core business is unsustainable as is. From what I understand, they use search APIs, which are expensive. So they need to get their own index, at least partial, to have a sustainable business.
Duck Duck Go uses the Bing API and they've been around for a long while. I'd imagine there's a heavy 80/20 rule of data that needs to be indexed frequently that people also search for, so those same cached results from API calls get the most use. Just speculation though.
Re: Kagi Sidekick (alpha)
#147Kagi founder here. This is very early (alpha) concept built by a single Kagi Labs developer in a few weeks. The proper infrastruture and product is not built yet. We are launching a prototype to get feedback and gauge demand. Why does this exist? It would be an efficient way for us to build and expand our own index. Assuming users of this would be Kagi users, we would expand our index by tens of thousands of high qua…
Because "AI" increases the shareholder's chances of winning the lottery.
Re: Kagi Sidekick (alpha)
#148Earlier quoted context omitted.
I have used Kagi for several months now, and I find that the ability to decide which searches I want quick answers for to be useful. There are certain searches where I feel comfortable with accepting a quick answer, and others where I don't think it would be useful. Being able to avoid the clutter unless I want it is nice.
Sports scores and stocks. If kagi inlined those results I would be very happy to not need to use the Google bang. Additionally, the google bang destroys suggested terms and with no kagi history I'm typing the whole query.
Re: bangs ruining suggestions: you can put the bang at the end of the query instead of at the beginning.
Re: Kagi Sidekick (alpha)
#149I've been a paying Kagi user for a few months now. The only thing that made me miss Google results was immediate answers: the answers that are extracted from a prominent web page and shown directly, so you don't have to click the link. Actually, Kagi has some of that for certain queries, but it's not as extensive as Google's. So, contrary to the most of the comments here, I support their AI endeavors for the sake of…
I have personally started to avoid Google’s and others immediate answers, because they are often very wrong because they are picked out of the contex. They try to show what you want to see, but it often means something else.
Re: Kagi Sidekick (alpha)
#150Earlier quoted context omitted.
Duck Duck Go uses the Bing API and they've been around for a long while. I'd imagine there's a heavy 80/20 rule of data that needs to be indexed frequently that people also search for, so those same cached results from API calls get the most use. Just speculation though.
I don't think it's a fair comparison. Every single other search engine sells ads including ddg. Sure, they don't track you, but will still serve ads for your quesries. Kagi is explicitly an ad free subscription. Comparing profitability is not straightforward.