5 ms·
At this point it should be clear that they know that Grok is unsafe to use, and will generate potentially illegal content even without a clear prompt asking it
by dleslie 9mo ago
At this point it should be clear that they know that Grok is unsafe to use, and will generate potentially illegal content even without a clear prompt asking it to do so.
This is a dangerous product, the manufacturer _knows_ it is dangerous, and yet still they provide the service for use.
- foxglacier 9mo ago[flagged]
- ribosometronome 9mo agoIt's hardly a hunt when there's bunch of warty green women with black tipped hats openly stirring a cauldron going "Double double toil and trouble." The world grew up and learned better. We've seen the effects that bullying, including cyberbullying, have on people. We've seen teenagers (and adults) get harassed with fake revenge porn.
- ebbi 9mo ago"pearl clutching nana" for simply stating CSAM is dangerous? I get the people arguing for free speech (in general) as there are lots of upsides. Where are the upsides in CSAM - whether real or computer generated? How does it benefit society?
- tredre3 9mo ago> Where are the upsides in CSAM - whether real or computer generated? How does it benefit society? In a public forum like X, there probably are no upsides. In general, though, pedophilia exists. This isn't something that is going to change. What is the harm in providing them with a alternative to real CSAM (which actually and actively hurts children)?
- mothballed 9mo agoGiving them no legal avenue allows us to put more of them in jail. Once they are in jail the odds of them molesting children goes from "possible very low but measurably above zero" to "~0." I think you are looking at it from the point of it being a punishment for victimizing someone -- when in fact it's used not to punish crime but to put away people who potentially might victimize someone in the future.
- techblueberry 9mo agoI actually think this is a profoundly interesting question and one that I'm interested in thinking about further. I think it's a problem for society when bad behavior is not transgressive. And moreover I'm less certain about this one, but I sort of think theoretically that society should be more liberal than it's institutions, and it creates really weird feedback loops when the institutions are more "liberal" than the population naturally. (I'm using the term generically, not directly aligned with the political meaning) I think the theory I would present is that people should not be encouraged to transgress further than they are impulsed to, but simultaneously people need an outlet to actually transgress in a way that is not acceptable! People shouldn't post edge memes because the algorithm encourages it. People should post edgy memes because it's transgressive! But when the institutions actually encourage it? How broken is it that you can't be an edgy teenager because edgy is the culture and not the counter-culture. In 2025, I think the truly transgressive activity is to not be online. Is to be straight-edge. And I sort of wonder if this is a small part of the young male mental-health crisis. They're not telling edgy jokes to be closer to their friends, they're telling edgy jokes to get fake internet points so people click on more advertising. How fucked is that? But it's like weird that kids are probably having less sex, drinking, smoking etc. then the institutions would have it. So to kind of answer your question, "In the 1990's, popular youth culture generally rebelled against this type of worry from adults but now even the youth are part of the moral witch-hunt for some reason." This might explain how I, a formerly "edgy" gen-x 90's kid am heartily against institutions supporting this kind of behavior, while simultaneously supporting people engaging in it. The adults; X, parents, etc. SHOULD be worrying about this kind of stuff SO THAT popular youth culture can continue to rebel against it.
- foxglacier 9mo agoCould you clarify what you mean by "institution"? What institution is actively encouraging transgression? Do you mean the cultures on social media that socially reward transgression? Isn't that just people and their culture, not an institution? Or is there something social media companies are actively doing to promote specific transgressions? I'm thinking about the 80s-90s worries about Christian heresy. Popular culture was (and still is) full of insults against Christianity, probably because that was the kind of thing that offended an older generation in the west at the time. Is it wrong for institutions to encourage that? While I have my own personal moral standards, I see society in general as morally relativist and don't accept arguments that the popular morals of today are right because they're popular now, while the popular morals of previous generations were wrong because they contradict the "right" morals of today. That's why I don't have much respect for people trying to enforce their own culture's arbitrary morals while not equally respecting conflicting morals. > when bad behavior is not transgressive That's a tricky one because what's "bad behavior"? Does it include denying the existence of God?
- GoatInGrey 9mo agoBoth of those things have and do motivate people to perform harm in the real world. It's not random beeps and boops that are nonsensical and challenging to manipulate behavior through.
- tzs 9mo ago> What's happened to the world where computer generated images and text are "dangerous". Do you think that images and text can ever be dangerous? If so, then why do you doubt that computer generated images and text can be dangerous? If not, then I'm not sure what the point of specifically mentioning computer generated.
- dleslie 9mo agoPossession of CSAM materials can be a criminal offence. Here in Canada, it doesn't matter if it's a sex doll that looks like a child, an imagined bit of erotica, or a drawing. And so, it is dangerous to use a tool that can inadvertently create such materials without user intent.
- foxglacier 9mo agoSo the danger comes from the government punishing people for possessing it? The underlying problem there is the law. Change that and AI will stop being dangerous.