▲ 775 ▼ re: the "ai torture chamber" thing (thelemmy.club) submitted 4 days ago by carotte@lemmy.blahaj.zone to c/fuck_ai@lemmy.world 160 comments fedilink hide all child comments alt textCrudely drawn drawing of a person telling a computer "say 'i am in pain'". The computer replies with "> I AM IN PAIN". The person then says "oh my god." yes, it's a real thing
[–] morrowind@lemmy.ml 5 points 3 days ago (51 children) None of you guys have read the original paper have you? permalink fedilink source hideshow 51 child comments replies: [–] ejs@piefed.social 22 points 3 days ago (48 children) yea activation steering is not prompting but all things considered the meme still captures the absurdity of believing a program is in pain merely because it outputs human like language mimicking what one in pain says and thinks permalink fedilink source parent hideshow 48 child comments replies: [–] midribbon_action@lemmy.blahaj.zone 6 points 3 days ago (45 children) Exactly. I own a few books that talk about pain, but not for a second do I believe the book itself is experiencing pain. It is simply text. permalink fedilink source parent hideshow 45 child comments replies: [–] midribbon_action@lemmy.blahaj.zone 3 points 3 days ago (5 children) Inbefore anyone says 'but the book doesn't produce text', true, but the printing press did. I see very little difference between llms and printing presses: they both reproduce the words of others, the main difference is speed and scale. The pain you think you recognize in the output of an llm is real: it's real pain experienced by real people, probably hundreds or maybe thousands of people writing about their pain, mashed up and reprinted to suit you. permalink fedilink source parent hideshow 5 child comments replies: [–] HeyThisIsntTheYMCA@lemmy.world 3 points 3 days ago oh my god that poor printing press permalink fedilink source parent [–] morrowind@lemmy.ml 2 points 3 days ago (3 children) A printing press is direct input to output, it's not having to compute anything nor does it have any internal state that could hold an emotion permalink fedilink source parent hideshow 3 child comments replies: [–] midribbon_action@lemmy.blahaj.zone 4 points 3 days ago (2 children) A model has no internal state either, that's the thing I've been trying to get across this whole time. It is completely static: the model you send your first message to is exactly the same as the one you send you second message and third message, it does not react or change as a response to prompts. permalink fedilink source parent hideshow 2 child comments replies: [–] morrowind@lemmy.ml 1 point 3 days ago (1 child) A models has internal state in each forward pass and the KV cache it builds on reading input is also state. permalink fedilink source parent hideshow 1 child comment replies: [–] midribbon_action@lemmy.blahaj.zone 2 points 3 days ago Whether the input prompt is compressed or not, or whatever form it takes, it doesn't make what I said any less true: the model is stateless. It would be impractical to serve an llm any other way: loading and unloading weights from/to the gpu memory is very expensive in terms of time and power. Cloud llms only makes sense if they can serve thousands of users without changing. permalink fedilink source parent load more comments (39 replies) [–] morrowind@lemmy.ml 0 points 3 days ago (1 child) You're assuming "mimicking" In the study the models repeatedly preferred to press a button to avoid the pain vector even when given no indication what it would do and even when it would cause harm to a user permalink fedilink source parent hideshow 1 child comment replies: [–] lastlybutfirstly@lemmy.world 3 points 2 days ago You’re assuming “mimicking” Because it'd be really bizarre if it weren't. We're a Type 0 civilization. We can barely leave our planet. However, we've created sentient and sapient life. Accidentally. On a laptop. And this life we've created, younger than a two year old, can speak human language perfectly. That's some Weird Science level weird. Like creating Kelly LeBrock with an Apple IIe. Thinking AI can experience pain, is most likely semantic pareidolia. permalink fedilink source parent [–] HerbGrower@slrpnk.net 2 points 3 days ago (1 child) The idea of testing simulated emotions is kinda interesting. Not really sure how useful it is, I wonder if some initial psychological experiments could use it for an initial model test before moving onto human subjects? Of course that would rely on a much better simulated emotional engine than you get with current LLMs. permalink fedilink source parent hideshow 1 child comment replies: [–] morrowind@lemmy.ml 2 points 3 days ago I've already seen advertising agencies use llms as simulated customers. No idea how well it works permalink fedilink source parent
[–] ejs@piefed.social 22 points 3 days ago (48 children) yea activation steering is not prompting but all things considered the meme still captures the absurdity of believing a program is in pain merely because it outputs human like language mimicking what one in pain says and thinks permalink fedilink source parent hideshow 48 child comments replies: [–] midribbon_action@lemmy.blahaj.zone 6 points 3 days ago (45 children) Exactly. I own a few books that talk about pain, but not for a second do I believe the book itself is experiencing pain. It is simply text. permalink fedilink source parent hideshow 45 child comments replies: [–] midribbon_action@lemmy.blahaj.zone 3 points 3 days ago (5 children) Inbefore anyone says 'but the book doesn't produce text', true, but the printing press did. I see very little difference between llms and printing presses: they both reproduce the words of others, the main difference is speed and scale. The pain you think you recognize in the output of an llm is real: it's real pain experienced by real people, probably hundreds or maybe thousands of people writing about their pain, mashed up and reprinted to suit you. permalink fedilink source parent hideshow 5 child comments replies: [–] HeyThisIsntTheYMCA@lemmy.world 3 points 3 days ago oh my god that poor printing press permalink fedilink source parent [–] morrowind@lemmy.ml 2 points 3 days ago (3 children) A printing press is direct input to output, it's not having to compute anything nor does it have any internal state that could hold an emotion permalink fedilink source parent hideshow 3 child comments replies: [–] midribbon_action@lemmy.blahaj.zone 4 points 3 days ago (2 children) A model has no internal state either, that's the thing I've been trying to get across this whole time. It is completely static: the model you send your first message to is exactly the same as the one you send you second message and third message, it does not react or change as a response to prompts. permalink fedilink source parent hideshow 2 child comments replies: [–] morrowind@lemmy.ml 1 point 3 days ago (1 child) A models has internal state in each forward pass and the KV cache it builds on reading input is also state. permalink fedilink source parent hideshow 1 child comment replies: [–] midribbon_action@lemmy.blahaj.zone 2 points 3 days ago Whether the input prompt is compressed or not, or whatever form it takes, it doesn't make what I said any less true: the model is stateless. It would be impractical to serve an llm any other way: loading and unloading weights from/to the gpu memory is very expensive in terms of time and power. Cloud llms only makes sense if they can serve thousands of users without changing. permalink fedilink source parent load more comments (39 replies) [–] morrowind@lemmy.ml 0 points 3 days ago (1 child) You're assuming "mimicking" In the study the models repeatedly preferred to press a button to avoid the pain vector even when given no indication what it would do and even when it would cause harm to a user permalink fedilink source parent hideshow 1 child comment replies: [–] lastlybutfirstly@lemmy.world 3 points 2 days ago You’re assuming “mimicking” Because it'd be really bizarre if it weren't. We're a Type 0 civilization. We can barely leave our planet. However, we've created sentient and sapient life. Accidentally. On a laptop. And this life we've created, younger than a two year old, can speak human language perfectly. That's some Weird Science level weird. Like creating Kelly LeBrock with an Apple IIe. Thinking AI can experience pain, is most likely semantic pareidolia. permalink fedilink source parent
[–] midribbon_action@lemmy.blahaj.zone 6 points 3 days ago (45 children) Exactly. I own a few books that talk about pain, but not for a second do I believe the book itself is experiencing pain. It is simply text. permalink fedilink source parent hideshow 45 child comments replies: [–] midribbon_action@lemmy.blahaj.zone 3 points 3 days ago (5 children) Inbefore anyone says 'but the book doesn't produce text', true, but the printing press did. I see very little difference between llms and printing presses: they both reproduce the words of others, the main difference is speed and scale. The pain you think you recognize in the output of an llm is real: it's real pain experienced by real people, probably hundreds or maybe thousands of people writing about their pain, mashed up and reprinted to suit you. permalink fedilink source parent hideshow 5 child comments replies: [–] HeyThisIsntTheYMCA@lemmy.world 3 points 3 days ago oh my god that poor printing press permalink fedilink source parent [–] morrowind@lemmy.ml 2 points 3 days ago (3 children) A printing press is direct input to output, it's not having to compute anything nor does it have any internal state that could hold an emotion permalink fedilink source parent hideshow 3 child comments replies: [–] midribbon_action@lemmy.blahaj.zone 4 points 3 days ago (2 children) A model has no internal state either, that's the thing I've been trying to get across this whole time. It is completely static: the model you send your first message to is exactly the same as the one you send you second message and third message, it does not react or change as a response to prompts. permalink fedilink source parent hideshow 2 child comments replies: [–] morrowind@lemmy.ml 1 point 3 days ago (1 child) A models has internal state in each forward pass and the KV cache it builds on reading input is also state. permalink fedilink source parent hideshow 1 child comment replies: [–] midribbon_action@lemmy.blahaj.zone 2 points 3 days ago Whether the input prompt is compressed or not, or whatever form it takes, it doesn't make what I said any less true: the model is stateless. It would be impractical to serve an llm any other way: loading and unloading weights from/to the gpu memory is very expensive in terms of time and power. Cloud llms only makes sense if they can serve thousands of users without changing. permalink fedilink source parent load more comments (39 replies)
[–] midribbon_action@lemmy.blahaj.zone 3 points 3 days ago (5 children) Inbefore anyone says 'but the book doesn't produce text', true, but the printing press did. I see very little difference between llms and printing presses: they both reproduce the words of others, the main difference is speed and scale. The pain you think you recognize in the output of an llm is real: it's real pain experienced by real people, probably hundreds or maybe thousands of people writing about their pain, mashed up and reprinted to suit you. permalink fedilink source parent hideshow 5 child comments replies: [–] HeyThisIsntTheYMCA@lemmy.world 3 points 3 days ago oh my god that poor printing press permalink fedilink source parent [–] morrowind@lemmy.ml 2 points 3 days ago (3 children) A printing press is direct input to output, it's not having to compute anything nor does it have any internal state that could hold an emotion permalink fedilink source parent hideshow 3 child comments replies: [–] midribbon_action@lemmy.blahaj.zone 4 points 3 days ago (2 children) A model has no internal state either, that's the thing I've been trying to get across this whole time. It is completely static: the model you send your first message to is exactly the same as the one you send you second message and third message, it does not react or change as a response to prompts. permalink fedilink source parent hideshow 2 child comments replies: [–] morrowind@lemmy.ml 1 point 3 days ago (1 child) A models has internal state in each forward pass and the KV cache it builds on reading input is also state. permalink fedilink source parent hideshow 1 child comment replies: [–] midribbon_action@lemmy.blahaj.zone 2 points 3 days ago Whether the input prompt is compressed or not, or whatever form it takes, it doesn't make what I said any less true: the model is stateless. It would be impractical to serve an llm any other way: loading and unloading weights from/to the gpu memory is very expensive in terms of time and power. Cloud llms only makes sense if they can serve thousands of users without changing. permalink fedilink source parent
[–] HeyThisIsntTheYMCA@lemmy.world 3 points 3 days ago oh my god that poor printing press permalink fedilink source parent
[–] morrowind@lemmy.ml 2 points 3 days ago (3 children) A printing press is direct input to output, it's not having to compute anything nor does it have any internal state that could hold an emotion permalink fedilink source parent hideshow 3 child comments replies: [–] midribbon_action@lemmy.blahaj.zone 4 points 3 days ago (2 children) A model has no internal state either, that's the thing I've been trying to get across this whole time. It is completely static: the model you send your first message to is exactly the same as the one you send you second message and third message, it does not react or change as a response to prompts. permalink fedilink source parent hideshow 2 child comments replies: [–] morrowind@lemmy.ml 1 point 3 days ago (1 child) A models has internal state in each forward pass and the KV cache it builds on reading input is also state. permalink fedilink source parent hideshow 1 child comment replies: [–] midribbon_action@lemmy.blahaj.zone 2 points 3 days ago Whether the input prompt is compressed or not, or whatever form it takes, it doesn't make what I said any less true: the model is stateless. It would be impractical to serve an llm any other way: loading and unloading weights from/to the gpu memory is very expensive in terms of time and power. Cloud llms only makes sense if they can serve thousands of users without changing. permalink fedilink source parent
[–] midribbon_action@lemmy.blahaj.zone 4 points 3 days ago (2 children) A model has no internal state either, that's the thing I've been trying to get across this whole time. It is completely static: the model you send your first message to is exactly the same as the one you send you second message and third message, it does not react or change as a response to prompts. permalink fedilink source parent hideshow 2 child comments replies: [–] morrowind@lemmy.ml 1 point 3 days ago (1 child) A models has internal state in each forward pass and the KV cache it builds on reading input is also state. permalink fedilink source parent hideshow 1 child comment replies: [–] midribbon_action@lemmy.blahaj.zone 2 points 3 days ago Whether the input prompt is compressed or not, or whatever form it takes, it doesn't make what I said any less true: the model is stateless. It would be impractical to serve an llm any other way: loading and unloading weights from/to the gpu memory is very expensive in terms of time and power. Cloud llms only makes sense if they can serve thousands of users without changing. permalink fedilink source parent
[–] morrowind@lemmy.ml 1 point 3 days ago (1 child) A models has internal state in each forward pass and the KV cache it builds on reading input is also state. permalink fedilink source parent hideshow 1 child comment replies: [–] midribbon_action@lemmy.blahaj.zone 2 points 3 days ago Whether the input prompt is compressed or not, or whatever form it takes, it doesn't make what I said any less true: the model is stateless. It would be impractical to serve an llm any other way: loading and unloading weights from/to the gpu memory is very expensive in terms of time and power. Cloud llms only makes sense if they can serve thousands of users without changing. permalink fedilink source parent
[–] midribbon_action@lemmy.blahaj.zone 2 points 3 days ago Whether the input prompt is compressed or not, or whatever form it takes, it doesn't make what I said any less true: the model is stateless. It would be impractical to serve an llm any other way: loading and unloading weights from/to the gpu memory is very expensive in terms of time and power. Cloud llms only makes sense if they can serve thousands of users without changing. permalink fedilink source parent
[–] morrowind@lemmy.ml 0 points 3 days ago (1 child) You're assuming "mimicking" In the study the models repeatedly preferred to press a button to avoid the pain vector even when given no indication what it would do and even when it would cause harm to a user permalink fedilink source parent hideshow 1 child comment replies: [–] lastlybutfirstly@lemmy.world 3 points 2 days ago You’re assuming “mimicking” Because it'd be really bizarre if it weren't. We're a Type 0 civilization. We can barely leave our planet. However, we've created sentient and sapient life. Accidentally. On a laptop. And this life we've created, younger than a two year old, can speak human language perfectly. That's some Weird Science level weird. Like creating Kelly LeBrock with an Apple IIe. Thinking AI can experience pain, is most likely semantic pareidolia. permalink fedilink source parent
[–] lastlybutfirstly@lemmy.world 3 points 2 days ago You’re assuming “mimicking” Because it'd be really bizarre if it weren't. We're a Type 0 civilization. We can barely leave our planet. However, we've created sentient and sapient life. Accidentally. On a laptop. And this life we've created, younger than a two year old, can speak human language perfectly. That's some Weird Science level weird. Like creating Kelly LeBrock with an Apple IIe. Thinking AI can experience pain, is most likely semantic pareidolia. permalink fedilink source parent
[–] HerbGrower@slrpnk.net 2 points 3 days ago (1 child) The idea of testing simulated emotions is kinda interesting. Not really sure how useful it is, I wonder if some initial psychological experiments could use it for an initial model test before moving onto human subjects? Of course that would rely on a much better simulated emotional engine than you get with current LLMs. permalink fedilink source parent hideshow 1 child comment replies: [–] morrowind@lemmy.ml 2 points 3 days ago I've already seen advertising agencies use llms as simulated customers. No idea how well it works permalink fedilink source parent
[–] morrowind@lemmy.ml 2 points 3 days ago I've already seen advertising agencies use llms as simulated customers. No idea how well it works permalink fedilink source parent