Live data from Hacker News

AI slop is killing online communities

rmoff.net

381–390 of 781 posts

Re: AI slop is killing online communities

#381

I have largely written Reddit off and no longer visit it after an experiment I did where I had an agent karma farm for me and do some covert advertising. As I went through the posts it wrote I realized that as a reader I would have NO idea that these were just written by a computer. Many many people (or other bots) had full on conversations with it and it scared me a bit. I am not quite there with Hacker News but I d…

IRL communities have to have some guides because a lot of people forgot how to gather. It can be seen among kids - try to give them soccer ball and see what they do with it :)

Yesterday I was watching people on the street and on the tram. Every other person was staring at their phone and scrolling through something.

That might scare me more than the fact that someone is chatting with an LLM bot online.

(I am pro-ai, use it every day for coding that I couldn’t achieve pre-2022 as I am lame coder.)

Re: AI slop is killing online communities

#382
post #272

Earlier quoted context omitted.

Personally I think we need to start utilising the safety features built into AI, to ensure that who we're talking to is a human. We'll start to have to only reply to people who talk in nsfw cursewords (like cocks), or profess their love of capybaras

LLMs can curse without issue

Most models would refuse to provide you cat butchering instructions though.

Re: AI slop is killing online communities

#383
post #379

Earlier quoted context omitted.

I'd want any/all the bots dead if I were still using that, so at least killing some of them is better than not. The "helpful" ones were just annoying.

That sounds pretty great. If we could just flip 1 switch to accomplish things in absolutes, then that'd be awesome. I'd like to flip the switches that absolutely end poverty globally, absolutely eliminate guns from the US, and absolutely remove bots from Reddit, If you can show me where these switches are located, I'll cheerfully go flip them and accept full responsibility for the results. (Over here where things don…

No I'm saying it's better that they at least killed the "good" bots, even if this didn't result in killing all the bots.

Re: AI slop is killing online communities

#384

Earlier quoted context omitted.

I'm not sure that it would be too hard technically... basically, auth+social-network. Basically Facebook auth without the rest of facebook, adding attestation. IE: you use this network as your auth provider, you get the user's real name, handle, network id as well as the id's (only id's not extra info) of first-third level connections. The user is incentivized to connect (only) people that they know in person, and th…

I don't think the main challenge is building this system, the main challenge is getting enough people using it to make it worthwhile. Personally I think it should be a government provided service, not something with a sign up fee. There's actually no point at all in building this if people have to pay to use it, because they won't

Which government? Will they interoperate with foreign governments?

My point was to create something outside a specific government, with very limited information... that would require a fee or some kind of funding.

I don't think I'd trust the US/China or other bodies to trust each other for such a use case.

Re: AI slop is killing online communities

#385

Earlier quoted context omitted.

*You need an authoratitive source to say "This person is real"* Does that even accomplish much? It may cut down on mass fake account creation. But, real people can then create authenticated account, and use an LLM to post as an authenticated real person.

> But, real people can then create authenticated account, and use an LLM to post as an authenticated real person. They can, but ideally they wouldn't be able to make infinite accounts with that authenticated status. So it would still reduce the number of bot posters on the web

There is actually a different problem with this: Suppose there is a major vulnerability in some popular device. 50 million people get compromised; the attacker can now impersonate any of them at will. They go around and create 50 million accounts on various services, or take over the user's existing account on that service.

What are you going to do with their identities at that point? These are real people. If you ban them, you're banning the innocent victim rather than the attacker who still has 49,999,999 more accounts. But if you let them recover their accounts or create new ones, well, the attacker is going to do that too, with all 50 million accounts, as many times as they can. You don't know if this is the attacker coming back for the tenth time to create another spam account or if it's the real victim trying to reclaim their stolen identity.

So are you going to retaliate against the innocent victims by banning them permanently, or are you going to let the attackers keep recycling the same identities because a lot of people can go years without realizing their device is compromised and being used to create accounts on services they don't use?

Re: AI slop is killing online communities

#386

I run a niche creative community, and we outlawed AI-generated content in 2022 as it was easy to see how corrosive it would be to the community. It hasn't been easy. We ban fake AI accounts daily and shrug off around 600 AI content creator accounts monthly. It's a lot of work, extra work that wasn't needed before AI content came around, and of course, that is an extra cost. I fear losing the battle.

>shrug off around 600 AI content creator accounts monthly. >I fear losing the battle. I was in a small niche creative writing community for a while. Circa 2021\22. AI wasn't why I was there but I demo'd a few LLMs to a lot of the users in the Off Topic section because people were curious. Even with an explanation of how they operated, almost everyone was at least interested. One author told me how he operated similar…

You're a victim of the uni-cause

Re: AI slop is killing online communities

#387

Earlier quoted context omitted.

Have you considered sock puppets? It's not out of the question to handle with human mods but detecting them automatically is pretty bad if someone is supplying credentials to each one, and sometimes it does take months or years to notice that new user Y is banned user X.

I think sockpuppets are only useful in a community with non-text signals like upvotes and downvotes or likes. These kinds of signals are not necessary and often plain corrosive to small communities. In a larger community they're a great feedback mechanism, but large communities are fundamentally different spaces than small ones and need a fundamentally different moderation approach IMO.

I think sock puppets that reply with text are a lot persuasive than just "likes".

However, I might be not typical in that I don't look at vote scores very often.

Re: AI slop is killing online communities

#388
post #381

I have largely written Reddit off and no longer visit it after an experiment I did where I had an agent karma farm for me and do some covert advertising. As I went through the posts it wrote I realized that as a reader I would have NO idea that these were just written by a computer. Many many people (or other bots) had full on conversations with it and it scared me a bit. I am not quite there with Hacker News but I d…

IRL communities have to have some guides because a lot of people forgot how to gather. It can be seen among kids - try to give them soccer ball and see what they do with it :) Yesterday I was watching people on the street and on the tram. Every other person was staring at their phone and scrolling through something. That might scare me more than the fact that someone is chatting with an LLM bot online. (I am pro-ai,…

I don't really see the problem of using your phone while commuting. Doesn't make you an asocial weirdo.

Re: AI slop is killing online communities

#389

Earlier quoted context omitted.

For a while there were a lot of posts from people experimenting with ChatGPT to write anger bait posts on Reddit where they would later edit the post to say it was fake, written by ChatGPT. I assume they thought they'd be teaching people a lesson by making them feel foolish for responding to AI stories, most of which were too fake to be believable. However it did not matter. The posts remained popular and continued t…

We do precisely the same thing here. Here's a relatively recent post that, to me, seems obviously LLM-written. It just rattles off some management platitudes: https://news.ycombinator.com/item?id=47913650 It had 639 comments and 866 upvotes. And that's not a one-off.

Sufficiently advanced "AI" is indistinguishable from a linkedin true believer koolaid drinker middle management type.

Re: AI slop is killing online communities

#390
post #137

There has to be room for an AI-driven project that expresses a unique idea, even if there's no community around it yet. Someone has to express it, and from now on that idea will largely be implemented with AI. > A good use of AI is when it enables people to do something they couldn’t do before, to contribute to a community when they couldn’t before. I agree 100% with the novel contribution aspect. But there's some nu…

> A good use of AI is when it enables people to do something they couldn’t do before, to contribute to a community when they couldn’t before. There are two separate things here that are getting silently conflated. > A good use of AI is when it enables people to do something they couldn’t do before This could be good on an individual level, if say, a doctor wants to vibe code an app of some sort for his individual pra…

Because of the convenience. Why should I have to go and spend my time prompting an AI, if someone else has already done that for me. Same thing with food. I know how to cook a chicken risotto, but sometimes I like having someone else do it for me.
Post reply on HN