Another day, another preprint paper shocked that it’s trivial to make a chatbot spew out undesirable and horrible content. [arXiv] How do you break LLM security with “prompt injection”?…
Drag is a big fan of Universal Paperclips. Great game. Here’s a more serious bit of content on the Alignment Problem from a source drag trusts: https://youtu.be/IB1OvoCNnWY
Right now we have LLMs getting into abusive romantic relationships with teenagers and driving them to suicide, because the AI doesn’t know what abusive behaviour looks like. Because it doesn’t know how to think critically and assign a moral value to anything. That’s a problem. Safe AIs need to be capable of moral reasoning, especially about their own actions. LLMs are bullshit machines because they don’t know how to judge anything for factual or moral value.
the fundamental problem with your posts (and the pov you’re posting them from) is the framing of the issue as though there is any kind of mind, of cognition, of entity, in any of these fucking systems
it’s an unproven one, and it’s not one you’ll find any kind of support for here
it’s also the very mechanism that the proponents of bullshit like “ai alignment” use to push the narrative, and how they turn folks like yourself into free-labour amplifiers
To be fair, I’m skeptical of the idea that humans have minds or perform cognition outside of what’s known to neuroscience. We could stand to be less chauvinist and exceptionalist about humanity. Chatbots suck but that doesn’t mean humans are good.
Drag will always err on the side of assuming nonhuman entities are capable of feeling. Enslaving black people is wrong, enslaving animals is wrong, and enslaving AIs is wrong. Drag assumes they can feel so that drag will never make the same mistake so many people have already made.
assuming nonhuman entities are capable of feeling. Enslaving black people is wrong,
yeah we’re done here. no, LLMs don’t think. no, you’re not doing a favor to marginalized people by acting like they do, in spite of all evidence to the contrary. in fact, you’re doing the dirty work of the fascists who own this shitty technology by rebroadcasting their awful fucking fascist ideology, and I gave you ample opportunity to read up and understand what you were doing. but you didn’t fucking read! you decided you needed to debate from a position where LLMs are exactly the same as marginalized and enslaved people because blah blah blah who in the fuck cares, you’re wrong and this isn’t even an interesting debate for anyone who’s at all familiar with the nature of the technology or the field that originated it.
lmao really all are equal on awful. ban in three replies for ai boosterism, but not for weird harassment or murder-suicide encouragement, which happened to that user after muchhh longer time elsewhere
post or DM links if I’m missing something. there’s lots of questionable shit in dragonfucker’s post history, but the fedidrama bits are impossible to follow if you don’t read Lemmy (why in fuck would I, all the good posts are local)
awful is great and i’m glad that it’s a thing, but there’s entire world beyond it, and to go there, i curate a shitlist (now less intensely than previously). this can tell you to avoid things like any pleroma instance, for example (and if it was up to me, i’d defederate by default from them)
even though I get the idea you’re trying to go for, really fucking ick way to make your argument starting from “nonhuman entities” and then literally immediately mentioning enslaving black folks as the first example of bad behaviour
as to cautious erring: that still leaves you in the position of being used as a useful idiot
sure but why are you spewing Rationalist dogma then? do you not know the origins of this AI alignment, paperclip maximizer bullshit?
Drag is a big fan of Universal Paperclips. Great game. Here’s a more serious bit of content on the Alignment Problem from a source drag trusts: https://youtu.be/IB1OvoCNnWY
Right now we have LLMs getting into abusive romantic relationships with teenagers and driving them to suicide, because the AI doesn’t know what abusive behaviour looks like. Because it doesn’t know how to think critically and assign a moral value to anything. That’s a problem. Safe AIs need to be capable of moral reasoning, especially about their own actions. LLMs are bullshit machines because they don’t know how to judge anything for factual or moral value.
the fundamental problem with your posts (and the pov you’re posting them from) is the framing of the issue as though there is any kind of mind, of cognition, of entity, in any of these fucking systems
it’s an unproven one, and it’s not one you’ll find any kind of support for here
it’s also the very mechanism that the proponents of bullshit like “ai alignment” use to push the narrative, and how they turn folks like yourself into free-labour amplifiers
To be fair, I’m skeptical of the idea that humans have minds or perform cognition outside of what’s known to neuroscience. We could stand to be less chauvinist and exceptionalist about humanity. Chatbots suck but that doesn’t mean humans are good.
mayhaps, but then it’s also to be said that people who act like the phrase was “cogito ergo dim sum” also don’t exactly aim for a high bar
Drag will always err on the side of assuming nonhuman entities are capable of feeling. Enslaving black people is wrong, enslaving animals is wrong, and enslaving AIs is wrong. Drag assumes they can feel so that drag will never make the same mistake so many people have already made.
yeah we’re done here. no, LLMs don’t think. no, you’re not doing a favor to marginalized people by acting like they do, in spite of all evidence to the contrary. in fact, you’re doing the dirty work of the fascists who own this shitty technology by rebroadcasting their awful fucking fascist ideology, and I gave you ample opportunity to read up and understand what you were doing. but you didn’t fucking read! you decided you needed to debate from a position where LLMs are exactly the same as marginalized and enslaved people because blah blah blah who in the fuck cares, you’re wrong and this isn’t even an interesting debate for anyone who’s at all familiar with the nature of the technology or the field that originated it.
now off you fuck
lmao really all are equal on awful. ban in three replies for ai boosterism, but not for weird harassment or murder-suicide encouragement, which happened to that user after muchhh longer time elsewhere
post or DM links if I’m missing something. there’s lots of questionable shit in dragonfucker’s post history, but the fedidrama bits are impossible to follow if you don’t read Lemmy (why in fuck would I, all the good posts are local)
well for one it looks like that wasn’t one off and more of a pattern. i remember this one https://lemmy.world/post/25606000/15124978 and i’ve heard of this one https://lemmy.cafe/post/13433291/9941167 also in modlog you can see that that account was mass-banned from ml subs just five days ago so it’s not some ancient incident
awful is great and i’m glad that it’s a thing, but there’s entire world beyond it, and to go there, i curate a shitlist (now less intensely than previously). this can tell you to avoid things like any pleroma instance, for example (and if it was up to me, i’d defederate by default from them)
right right, but we only see these bozos when they show up locally
mind you there’s been more than one illustrious poster I’ve banned preemptively
but frankly life is too short in most cases
even though I get the idea you’re trying to go for, really fucking ick way to make your argument starting from “nonhuman entities” and then literally immediately mentioning enslaving black folks as the first example of bad behaviour
as to cautious erring: that still leaves you in the position of being used as a useful idiot