Change Reddit so that if 3 people tag your post with an "AI" tag, you have to do a captcha before you can post again.

[-]

starrymochiie@reddit

That would turn every post into a group report system real quick “3 friends didn’t like it “ = gone forever is wild stuff

[-]

It doesn't mean "gone forever" it means you have to do a captcha to post again. If you're a human, that's not a hardship, at most it's a mild annoyance. If you're not a human, it might stop you from posting until the human that set you up checks in again, it could take you out of spamming for the day or longer.

[-]

Baerog@reddit

As someone who often posts against the Reddit hivemind, this would be a nightmare. I'd be doing captchas on every single post. Honestly, even non-controversial posters would. 3 people is nothing, you could find 3 people that dislike almost any single post.

[-]

limbodog@reddit (OP)

Nah. Once you do a captcha you're immune for hours

[-]

ThePython11010@reddit

They'll just do the captcha, then turn on the bot and spam a few dozen slop posts over the next few hours.

[-]

Forsaken_Counter_887@reddit

And why should I have to do that just because someone didn't like my opinion?

[-]

limbodog@reddit (OP)

Because everything else reddit is doing is failing

[-]

Forsaken_Counter_887@reddit

So just make Reddit judge every comment automatically? Only doing it sometimes, as a result of people's opinions, makes it guaranteed to be abused for purposes it wasn't intended for.

[-]

limbodog@reddit (OP)

I have no idea how you reached either of those conclusions

[-]

Forsaken_Counter_887@reddit

That's what worries me

[-]

limbodog@reddit (OP)

There's no need to judge every comment, and it is based on whether a comment appears to be AI.

[-]

xrelaht@reddit

Humans are terrible at spotting AI generated text. This would do nothing except inconvenience people.

[-]

limbodog@reddit (OP)

Nah.

First, that's why it takes more than one human to do it. Second, that's why it's only a minor inconvenience for a human redditor. Third it's crowdsourcing, it might not cure the problem, but it can make it hard enough to reduce the number of bot and AI posts (which is extreme)

[-]

xrelaht@reddit

Multiple people doing something they’re bad at doesn’t make it work better. And since CAPTCHAs are also no problem for bots at this point, you really would just be inconveniencing people for no reason.

But since you’re so dug in about how great this would be, it doesn’t belong here and you should delete it.

[-]

limbodog@reddit (OP)

There are more advanced captchas and cloudfire tests now.

And multiple people doing something they aren't great at does, in fact, make it work better. Thats' why crowdsourcing works.

And if you're this bothered by unusual ideas you should unsubscribe from this subreddit. It's not for you.

[-]

GrouchyResearcher392@reddit

To prove you’re an actual human being

[-]

Forsaken_Counter_887@reddit

And why should I have to do that just because someone didn't like my opinion?

[-]

GrouchyResearcher392@reddit

So the websites backend can verify that that’s an actual opinion that you as a person hold and aren’t a propaganda bot designed to sow malice and discord among the population.

[-]

Forsaken_Counter_887@reddit

I can't tell if you're being intentionally obtuse or if it's just accidental. Let me try a third time:

And why should I have to do that JUST BECAUSE SOMEONE DIDN'T LIKE MY OPINION?

[-]

Lele92007@reddit

It's not about the content of the post. Reposts of top all time pictures and such would get flagged generally rightfully as they are from karma farming bots. It's a much greater annoyance to bot farms than to humans.

[-]

GrouchyResearcher392@reddit

Oh, are we big typing now?

BECAUSE THE INTERNET IS RAMPANTLY FILLED WITH BOTS THAT POST THE MOST MINDLESS UNNEEDED STUFF IMAGINABLE AND IF FUCKING TYPING SOME LETTERS AND CLICKING ON STOPLIGHTS IS TOO FUCKING HARD FOR YOU TO ACCOMPLISH THEN WE WILL GLADLY SACRIFICE YOUR POSTS FOR A BOT FREE INTERNET.

[-]

w3woody@reddit

Sometimes my wife uses OpenAI to pass captcha tests because she's bad at them.

[-]

Burger_Destoyer@reddit

How often are you making Reddit posts anyway? 9/10 times the only people posting more than once a day are karma farming bots/accounts.

[-]

joe_canadian@reddit

I'm autistic and work in legal - I've been accused of being AI way too much. This would annoy the hell out of me.

[-]

limbodog@reddit (OP)

It'll probably annoy a lot of people a little bit having to do a captcha every few hours assuming you were on reddit 12 hours a day.

But on the other hand it will reduce the bot traffic, which is extreme, and that's plenty annoying too

[-]

Tensor3@reddit

You are aware that most modern bots have no problem with captchas, right? Captchas are for ai model training, not keeping bots out

[-]

limbodog@reddit (OP)

Cloudfire, Captchav3, etc. The defenses are being upgraded at the same time as the AIs. It will at least keep out the less advanced ones. And when new tech comes out, you switch to that. If we get to the point that there's no way to challenge an AI then the internet will be useless and it won't matter.

[-]

theonetruegrinch@reddit

Then how will reddit make money by helping train AI ?

[-]

Tensor3@reddit

Captches main purpose is to train ai

[-]

Traditional-Hall-591@reddit

That would never be weaponized.

[-]

twinklepocket@reddit

People would start tagging anything they dislike as AI.

[-]

limbodog@reddit (OP)

You just limit how often it can take effect.

[-]

YrdoomZ@reddit

Sounds great, as long as there is some punishment for the accounts that used the AI tag if the reported poster passes the Captcha so it can’t be weaponized, like a week long ban from Reddit or something

[-]

JoshuaSuhaimi@reddit

there are humans who copy and paste AI responses though

[-]

GrouchyResearcher392@reddit

Bot by proxy

[-]

JoshuaSuhaimi@reddit

those people would be able to solve captchas is my point

[-]

GrouchyResearcher392@reddit

Yeah that’s true too, didn’t think of that. I feel like, besides the most blatantly brain dead though, people would start just typing down the words they are thinking more often rather going to chat got and prompting some “make a response to this comment that says “xxxyyz”

[-]

limbodog@reddit (OP)

No need for that. Just limit how often it can be done to someone

[-]

Jazzlike-Wolf6265@reddit

Mass reporting made easy. Even a random person could run 3 accounts to temporarily block someone they don't like.

[-]

limbodog@reddit (OP)

It would block them for about 10 seconds

[-]

Fa1nted_for_real@reddit

I run 3 accounts just because lol.

(Main, business, and secondary)

[-]