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

SlopStop: Community-driven AI slop detection in Kagi Search

blog.kagi.com

Post reply on HN
Print view View original

241–250 of 271 posts

Re: SlopStop: Community-driven AI slop detection in Kagi Search

#241

Post by spencerflem »

chromehearts wrote: post #237
czottmann wrote: post #235

Earlier quoted context omitted.

What makes you think this is slop? Also, I think many people use the term "slop" and "AI was involved" interchangeably, but to me, they're not synonymous. To me, writing blog posts with the help of AI is fine (grammar checks, structural help etc.) while auto-generated content generation w/o human oversight is not.

The negated sentence structure "X isn't just Y -- it's Z" directly followed by a list of 3 or 4 bullet points. Maybe the bullet points are a heavy reach but nobody can tell me otherwise of the former. I agree on your first part! The whole article does read like slop tho; it's more like "Human was involved" here

AI writes like that because it was trained on the internet, which by now is mostly marketing copy.

Re: SlopStop: Community-driven AI slop detection in Kagi Search

#242

Post by pseudalopex »

jacquesm wrote: post #219
pseudalopex wrote: post #197

Earlier quoted context omitted.

> It's still crap, but it outperforms the real thing of look and feel as well as general language skills while it underperforms in the part that matters. The real thing meant human SEO spam? Or human writing?

Actual knowledge and understanding.

> It's still crap, but it outperforms Actual knowledge and understanding of look and feel as well as general language skills while it underperforms in the part that matters.

I would think this meant AI generated pages performed better in search engines but looked, felt, used language, and informed worse. But you said the crap looks good.

Re: SlopStop: Community-driven AI slop detection in Kagi Search

#243

Post by Schiendelman »

VHRanger wrote: post #72
chickensong wrote: post #68

> Our review team takes it from there How does this work? Kagi pays for hordes of reviewers? Do the reviewers use state of the art tools to assist in confirming slop, or is this another case of outsourcing moderation to sweat shops in poor countries? How does this scale?

Hey, Kagi ML lead here. > Kagi pays for hordes of reviewers? Is this another case of outsourcing moderation to sweat shops in poor countries? No, we're simply not paying for review of content at the moment, nor is it planned. We'll scale human review as needed with long time kagi users in our discord we already trust > Do the reviewers use state of the art tools to assist in confirming slop Mostly this, yes. For imag…

Thank you for jumping in with such a detailed response - and for the work you do! You're making the internet a little better.

Re: SlopStop: Community-driven AI slop detection in Kagi Search

#245

Post by jadbox »

rrr_oh_man wrote: post #226
jacquesm wrote: post #81

HN could use some of this. It'd be nice if there was a safe having from the equivalent of high grade junk mail.

I built https://itter.sh because of that

Is there any metrics on usage/users/posts? Lovely project but is it more productive that just echoing to /dev/null?

Re: SlopStop: Community-driven AI slop detection in Kagi Search

#246

Post by bayindirh »

chemotaxis wrote: post #205
postalcoder wrote: post #142

Earlier quoted context omitted.

I understand the substack exclusion. The paywall is not user friendly. If you don't mind, it'd be cool to take a look at your bookmark domains so that I could potentially augment the filter on my site. If you're interested, my email is in bio.

There's no paywall on Substack, except for actual paid-only blogs. You might be thinking Medium.

Substack has a very disconcerting reading experience.

I generally close the tab when the inevitable "Subscribe for the newsletter. Almost half of the world has subscribed, c'mon!" overlay appears.

I just want to read the thing, not to fill my inbox with newsletters from various blogs. We have RSS for that!

Re: SlopStop: Community-driven AI slop detection in Kagi Search

#247

Post by baggachipz »

peddling-brink wrote: post #192
vachina wrote: post #187

Earlier quoted context omitted.

Thing is LLM can be trained on things not available on the public internet, unlike search engines having to return public URLs. I’m sure all these agentic AI are slurping in all proprietary codebases and their documentations and training on them. It’s the only way to one up the competition.

That is the part which most frightens me. If the LLMs become the bastions of truth because the open web has fallen to slop, truth becomes privatized and unverifiable.

And hallucinated a significant percentage of the time.

Re: SlopStop: Community-driven AI slop detection in Kagi Search

#248

Post by righthand »

roman_soldier wrote: post #213
righthand wrote: post #181

Earlier quoted context omitted.

You must not use Kagi because a "human slop" system is available on both Kagi and HN. It's called a downvote and the article has an image how you can downvote links in search results. Just an FYI why you're getting downvoted for posting a dire comment yourself.

Then people can downvote AI slop or Human slop equally. Why do we need to discriminate against digital intelligence which is often leagues above the average mouth breathing Joe.

Because digital intelligence isn’t leagues above the average mouth breathing Joe. With out the Joe there is zero digital intelligence. Llms being trained on the internet and a bunch of media aren’t the smartest brain in the world. It’s just compression of intelligence and a lot of media is trash. Likely Llms are dumber than the average Joe just more well read.

Re: SlopStop: Community-driven AI slop detection in Kagi Search

#249

Post by thefringthing »

postalcoder wrote: post #134
irl_zebra wrote: post #80

This is so, so exciting. I hope HN takes inspiration and adds a similar flag. :)

I just requested access to the database @freediver so hopefully it should be integrated into https://hcker.news soon. I appreciate Kagi's community-driven approach. The open Small Web list[0] is invaluable. Applying a smallweb filter[1] on HN brings a breath of fresh air to the frontpage. 0: https://github.com/kagisearch/smallweb 1: https://hcker.news/?smallweb=true

Is there a simple way to turn a set of hcker.news feed settings into an RSS feed?

Re: SlopStop: Community-driven AI slop detection in Kagi Search

#250

Post by freediver »

pseudalopex wrote: post #41
Zambyte wrote: post #31

Earlier quoted context omitted.

Been using Kagi for two years now. Their consistent approach to AI is to offer it, but only when explicitly requested. This is not that surprising with that in mind.

> Their consistent approach to AI is to offer it, but only when explicitly requested. Kagi News does not disclose AI even.

The code is open source, you can add it as a PR as you see appropriate.
Post reply on HN