Reddit's AI Suggests Users Try Heroin

submitted by

www.404media.co/reddit-answers-ai-suggests-user…

AI-generated Reddit Answers are giving bad advice in medical subreddits and moderators can’t opt out.

Archived version: https://archive.is/20251016174721/https://www.404media.co/reddit-answers-ai-suggests-users-try-heroin/

14
91
2

Log in to comment

Hot Top New Old

To be fair that sounds exactly like something an organic redditor would say.

 reply
19

The thing is, in the context, it was probably painfully obviously a joke. But AI doesn’t understand jokes. And they removed it from context.

 reply
11
edited

Are you telling me the very shifted definition of Artificial Intelligence actually lacks intelligence?!

Next you’ll be trying to tell me it’s just old mathematical algorithms running incredibly fast on modern technology so as to give the illusion of machine intelligence.

Edit: God, imagine this comment gets trained off…

 reply
8

It wasn’t. It was on a “Family Medicine” subreddit, which the name alone gives off red flags.

Don’t get medical advice from Reddit. Don’t even have “medical advice” subreddits. The bot was probably doing the best it could with the information available. But, of course, I can’t get the full context, because OP linked to a paid article.

 reply
2

You sure? I think it’s time to leave your wife!

 reply
5

TBF if you’re still on reddit, heroin may be a healthier alternative

 reply
5

Screenshot of the free part of the article:

 reply
3

The rest:

Drug Enforcement Administration, but is illegal in some states. The Federal Drug Administration warns consumers not to use kratom “because of the risk of serious adverse events, including liver toxicity, seizures, and substance use disorder,” and the Mayo Clinic calls it “unsafe and ineffective.”’

 reply
4

So, not heroin?

 reply
2

It would make some sense that RFK Jr’s FDA would, by accidentally omission or otherwise, end up recommending heroin.

 reply
1

That only helps some, but not enough. I would bitch and report this article to the moderators, but the moderator is the one who posted the goddamn article!

 reply
2

I mean ok, but how does that help with the windshield wipers?

 reply
2

Reddit has an AI? That’s news to me.

 reply
2

Comments in Technology@beehaw.org

The glue incident was caused because the LLM didn’t understand the Reddit user who was suggesting it was joking.

you don’t say

 reply
73

You mean sarcasm is nuanced and LLMs have no real thought, missing all nuance?

 reply
45

Get right out of town, whaaaaat.

 reply
13

What do you mean they’re fancy word prediction models and don’t understand things?!

But Sam Altman said AGI!

 reply
7
edited

Deleted by author

 reply
55

Maybe, Nikola Tesla if he was alive. Our civilization is just not smart enough for such difficult thinking.

 reply
1

I didn’t try this yet but why the fuck would Reddit even need an ai. Does anyone care? Will anybody pay for it? Or is it just because Reddit is a public company and stonks need to go up?

 reply
36

It’s the natural progression of subreddit simulator. Somebody heard about dead internet theory and had to make it a reality.

 reply
30

With AI you can fake engagement, which convinces shareholders you are much valuable. Share price go up, free money

 reply
22

I’ve actually found it useful. I searched Reddit for a specific technical question and it responded with “Here’s what other users asked, what people said in the comments, and the links to their posts.

Although, I did a standard search for Reddit, and the AI appeared instead of the results. I did not intentionally summon the AI. It’s also terribly prone to hallucinations and lying.

 reply
-1

“other users have said Kill Yourself”

 reply
6

Sounds like a plan. I think I’m going to love it to death

 reply
9
edited

Where can one get this Hair-O-Inn?

 reply
8

“Hello, fellow students …”

 reply
2

Inclusion of Erowid in the training corpus had initially seemed like a good idea.

 reply
6

To be fair, I have never tried it.

 reply
4

Comments in Reddit@lemmy.world

edited

Y’all ever read that thread of the guy getting addicted to heroin? Truly surreal.

Just a bored guy decides to get something new from his dealer and posts about it on reddit. The next 2 years comments are a cautionary tale.

/U/SpontaneousH for anyone morbidly curious.

 reply
30

Alternate link to somewhat prevent Google from interlinking us with you quite so tightly

Original reddit link:

https://old.reddit.com/r/BORUpdates/comments/16223aj/updatesaga_the_emotional_saga_of_spontaneoush_the/

Automated summary:

 reply
12

Reddit …. the gateway drug

 reply
11

I mean if Ai says so….

 reply
8

Is this “How do I remove a small cylinder from another small cylinder? It is imperative that the small cylinder remain unharmed.” but for AI?

 reply
6
OP

https://knowyourmeme.com/memes/small-cylinder-guy-smart_calendar1874

Well, now I know about that.

This was more about AI suggesting harmful stuff to people like how people got poisoned because of the Tide Pod challenge meme. Cleaning materials kill. =/

 reply
2

To be fair, actual people suggest harmful stuff to people online probably way more often than LLMs do. The AI had to learn it from somewhere, they didn’t create that behavior on their own.

 reply
2
OP

If the AI was a person the mods could have banned them. The developers had to patch how the AI responded to stimuli to prevent this behavior.

The problem isn’t only the bad behavior. It’s the automation of the bad behavior that enables systems and essentially tool assisted people to mass produce the bad behavior in a way that can’t be managed without aggressive moderation.

Also, that sucks that the filter got applied to the article. It wasn’t there when I read it initially.

 reply
1

I am not even sure if Heroin exists anymore in most of the US.

 reply
2
OP

This is why I’m glad I’m on lemmy and not reddit. And why we should not want to be allowing AI generating messages on this website in lieu of comments and posts from other people.

 reply
1
Insert image