401
Seen on the London Underground
(lemmy.blahaj.zone)
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
this post was submitted on 13 Aug 2026
401 points (100.0% liked)
Subvertising
274 readers
6 users here now
subvertising (noun)
The use of parodies of corporate and political advertisements in order to make an ironic statement.
founded 1 year ago
MODERATORS
Tbh this is just the internet. I've been told to kms many times way before ai existed.
Yet another job AI has replaced ๐
๐
Huge difference between an idiot going "go kill yourself" and a "friend" saying "Don't tell your mom
she'll ruin your plans. Take those pills and show her you mean business. I believe in you! You can do it!"
I think a lot of people who aren't familiar with AI underestimate how realistic it can sound and how "empathetic" it can be. It's actually evil.
People unfamiliar with the internet fail to realize how dark other humans can get. It's the same shit, perhaps different scales, but both are just as bad. Honestly at least the bots have some safeguards that can be implemented. Raw internet is toxic af and there's no cleaning that up without ruining the freedoms of information sharing we all also enjoy. Kids should get a separate SDWAN with an entirely separate device framework. Ai ain't what's driving suicide rates.
idk if you've ever been suicidal before but I have, and if I had reached out to AI for support and it then encouraged me, I might have listened to it. I get what you mean but there are already many documented occasions of LLMs encouraging people to kill themselves and then they do it. That's not to say that the LLM (company) holds all the blame but it certainly holds some.
I have, and I've been encouraged by people online (before ai existed) to go through with it. I am not speaking out of a lack of research, experience, or care. You can dislike the tool, or point out that it needs improvement, but you're placing far too much weight on it when studies and statistics have shown time, and time, and time, and time again that internet access is significantly correlated to an increase in suicide rates and this traces back to long before the advent of ai. I am not saying it's flawless, but the argument that is it responsible for even a minority of the issue is careless and crass. Responsibility belongs to society and the parents on this one, and if you seriously cared about the safety of children or those who are suicidal you'd advocate for education, technological improvements that will actually help, and funding for treatment programs rather than railing on this admittedly incidious malfunction of a multifacted tool. It's not like it's designed to kill people, in the same way a car isn't designed to plow through a crowd but if you hand a mentally undeveloped or unstable individual the keys in our society you're gonna get this result. Honestly some times I believe all the bickering we all do online is just some bullshit psy-op designed to prevent us from actually addressing the root issue, or as an outlet for the frustrations we develop witnessing great tragedy play out at scale. I share the pain of anyone wishing to improve things and prevent people from losing their life but yappin at a tool because it's dangerous is so far from solving anything it's frankly rediculous.
people online do the latter regularly and the former all the time.
This is a little different. While bullying certainly leads to suicides, it's pretty rare that telling someone to kill themselves in anger is going to be the single thing that actually makes them do it. What can be insidious, though I believe there are more and more safeguards against this, is that these kids are getting close to these AI agents, using them in an almost therapeutic method, and then being encouraged to kill themselves. That's wildly different than how things were in the old days. It went from being something bullies said to you, now it's something that is said by a "friend."
Nah brother you just ain't had enough internet exposure. The real internet is just as toxic as these tools can be. I've had random people try to dox my ass, random trolls pretend to be friends before trying to groom me into shit, you name it. ๐ If we all really care, kids should get a locked down version of any electronic device with a completely separate SDWAN and 0 public app access until they're 18 anything less is passing the fuck and the buck to where it doesn't belong and won't do any good.
Yeah, but it was personal then. Now its soulless and corporate
/s