Every technology is a tool - both safe and unsafe depending on the user.
Nuclear technology can be used to kill every human on earth. It can also be used to provide power and warmth for every human.
AI is no different. It can be used for good or evil. It all depends on the people. Vilifying the tool itself is a fool’s argument that has been used since the days of the printing press.
My big problems with AI are the climate cost and the unethical way that a lot of these models have been trained. If they can fix those, then yeah I don’t have an issue with people using it when it’s appropriate but currently lots of people are using it out of sheer laziness. If corpos are just using it to badly replace workers and kids are using it instead of learning how to write a fucking paragraph properly, then yeah, I’ll hate on AI
While this may be true for technologies, tools are distinctly NOT inherently neutral. Consider the automatic rifle or the nuclear bomb. In the rifle, the technology of the mechanisms in the gun is the same precision-milled clockwork engineering that is used for worldwide production automation. The technology of the harnessing of a nuclear chain reaction is the same, whether enriching uranium for a bomb or a power plant.
HOWEVER, BOTH the automatic rifle and the nuclear bomb are tools, and tools have a specific purpose. In these cases, that SOLE purpose is to, in an incredibly short period of time, with little effort or skill, enable the user to end the lives of as many people as possible. You can never use a bomb as a power plant, nor a rifle to alleviate supply shortages (except, perhaps, by a very direct reduction in demand). Here, our problem has never been with the technology of Artificial Neural Nets, which have been around for decades. It isn’t even with “AI” (note that no extant “AI” is actually “intelligent”)! No, our problem is with the tools. These tools are made with purpose and intent. Intent to defraud, intent to steal credit for the works of others, and the purpose of allowing corporations to save money on coding, staffing, and accountability for their actions, the purpose of having a black box a CEO can point to, shrug their shoulders, and say “what am I supposed to do? The AI agent told me to fire all of these people! Is it my fault that they were all <insert targetable group here>?!”
These tools cannot be used to know things. They are probabilistic models. These tools cannot be used to think for you. They are Chinese Rooms. For you to imply that the designers of these models are blameless — when their AI agents misidentify black men as criminals in facial recognition software; when their training data breaks every copyright law on the fucking planet, only to allow corporations to deepfake away any actual human talent in existence; when the language models spew vitriol and raging misinformation with the slightest accidental prompting, and can be hard-limited to only allow propagandized slop to be produced, or tailored to the whims of whatever despot directs the trolls today; when everyone now has to question whether they are even talking to a real person, or just a dim reflection, echoing and aping humanity like some unseen monster in the woods — is irreconcilable with even an iota of critical thought. Consider more carefully when next you speak, for your corporate-apologist principles will only help you long enough for someone to train your beloved “tool” on you. May you be replaced quickly.
You’ve made many incorrect assumptions and setup several strawmen fallacies. Rather than try to converse with someone who is only looking to feed their confirmation bias, I’ll suggest you continue your learnings by looking up the Dunning Kruger effect.
Can you point out and explain each strawman in detail? It sounds more like someone made good analogies that counter your point and you buzzword vomited in response.
Dissecting his wall of text would take longer than I’d like, but I would be happy to provide a few examples:
I have “…corporate-apologist principles”.
— Though wolfram claims to have read my post history, he seems to have completely missed my many posts hating on TSLA, robber barons, Reddit execs, etc. I completely agree with him that AI will be used for evil by corporate assholes, but I also believe it will be used for good (just like any other technology).
“…tools are distinctly NOT inherently neutral. Consider the automatic rifle or the nuclear bomb”
“HOWEVER, BOTH the automatic rifle and the nuclear bomb are tools, and tools have a specific purpose”
— Tools are neutral. They have more than one purpose. A nuclear bomb could be used to warm the atmosphere another planet to make it habitable. Not to mention any weapon can be used to defend humanity, or to attack it. Tools might be designed with a specific purpose in mind, but they can always be used for multiple purposes.
There are a ton of invalid assumptions about machine learning as well, but I’m not interested in wasting time on someone who believes they know everything.
I understand that you disagree with their points, but I’m more interested in where the strawman arguments are. I don’t see any, and I’d like to understand if I’m missing a clear fallacy due to my own biases or not.
Many of their points are factually incorrect. The first point I refuted is a strawman argument. They created a position I do not hold to make it easier to attack.
EDIT: now I understand. After going through your comments, I can see that you just claim confirmation bias rather than actually having to support your own arguments. Ironic that you seem to show all of this erudition in your comments, but as soon as anyone questions your beliefs, you just resort to logical buzzwords. The literal definition of the bias you claim to find. Tragic. Blocked.
Blocking individual on Lemmy is actually quite pointless as they still can reply to your comments and posts you just will not know about it while there can be whole pages long slander about you right under your nose
I’d say it’s by design to spread tankie propaganda unabated
You know what? They can go ahead and slander me. Fine. Good for them. They’ve shown they aren’t interested in actual argument. I agree with your point about the whole slander thing, and maybe there is some sad little invective, “full of sound and fury, signifying nothing”, further belittling my intelligence to try to console themself. If other people read it and think “yeah that dude’s right”, then that’s their prerogative. I’ve made my case, and it seems the best they can come up with is projection and baseless accusation by buzzword. I need no further proof of their disingenuity.
Least you ought to care about is who thinks what about your comments. I mean I don’t wanna tell you what to think but it seems sane to just ignore that
This forum means nothing, nothing we or anyone says here is of any importance. It’s not even something that is saved anywhere in the long run but is merely some speck of digital dust that will fly away as soon as you turn off your phone
It’s the most meaningless words you will ever write, here
So you know, you either enjoy drama or memes or some arcane knowledge or what is exactly you are doing on the internet dear stranger? I hope you aren’t thinking what you say here has any meaning because no one cares and I can assure you I won’t remember about any of this in 10 minutes tops
I feel like you are trying to say something serious here and I assure you this is inherently absurd. Whatever you wanna change or whatever you want to cope with start outside the phone/tablet/laptop
A take with which I disagree, but mostly a fair one. If you want to live an unexamined life, it’s not like that’s any of my business. You are absolutely correct on your read of me. I am committed to arcane esoterica, so that’s what I’m doing here. I’m here to discuss, and to learn, and I genuinely want to explore the limits of my understandings, identify my biases, and catch my fallacies, so when someone throws out some take that seems spectacularly inane, like “all technologies are tools, and all tools are the same”, I want to know if I’m missing something.
As far as why I dare to be so “absurd”: I’m inclined to wonder why you have decided that your personal nihilism means that no-one else should be able to have a serious discussion of any kind. That’s kind of sad.
You are on the internet how do you want to have serious discussions when you don’t even know who are you talking to. I may be a toddler for all you know it or some kind of psychotic serial killer chilling with a cigarette after the job well done. Like seriously people get the heck out of social media and do something actually that matters
How am I supposed talk seriously when you call my common sense a nihilism too lol. Just focus your capacity of seriousness for real life, thats the only sane way to go around. Or feel free to go insane on the internet discussing geopolitics with toddlers and ai and drunk college dropouts.
Enjoy getting baited every 5 seconds and probably be angry and miserable daily because you treat your phone with funny digital people so firkin seriously
Even this is getting too serious lol, like cmon I can’t believe I am actually doing this and every fiber of my being recoils right now saying “run, there’s nothing productive here, get the frick away and rain missiles from afar instead of getting involved in the pointless trenches of internet keyboard warfare”
Blocking means that you don’t have to devote your time and thoughts to that person. That’s pretty valuable.
And even if they decide they are going to attack you, not-responding is often a good strategy vs that kind of crap anyway - to avoid getting pulled into an endless bad-faith argument. (I’d still suggest not announcing that you’ve blocked them though. Just block and forget about it.)
Every tech can be safe and unsafe? I think you’ve oversimplified to the point of meaninglessness. Obviously some technologies are safer than others, and some are more useful than others, and some have overwhelming negative effects. Different tech can and should be discussed and considered on a case by case basis - not just some “every tech is good and bad” nonsense.
Every technology is a tool - both safe and unsafe depending on the user.
Nuclear technology can be used to kill every human on earth. It can also be used to provide power and warmth for every human.
AI is no different. It can be used for good or evil. It all depends on the people. Vilifying the tool itself is a fool’s argument that has been used since the days of the printing press.
My big problems with AI are the climate cost and the unethical way that a lot of these models have been trained. If they can fix those, then yeah I don’t have an issue with people using it when it’s appropriate but currently lots of people are using it out of sheer laziness. If corpos are just using it to badly replace workers and kids are using it instead of learning how to write a fucking paragraph properly, then yeah, I’ll hate on AI
Been this way since the harnessing of fire or the building of the wheel.
While this may be true for technologies, tools are distinctly NOT inherently neutral. Consider the automatic rifle or the nuclear bomb. In the rifle, the technology of the mechanisms in the gun is the same precision-milled clockwork engineering that is used for worldwide production automation. The technology of the harnessing of a nuclear chain reaction is the same, whether enriching uranium for a bomb or a power plant.
HOWEVER, BOTH the automatic rifle and the nuclear bomb are tools, and tools have a specific purpose. In these cases, that SOLE purpose is to, in an incredibly short period of time, with little effort or skill, enable the user to end the lives of as many people as possible. You can never use a bomb as a power plant, nor a rifle to alleviate supply shortages (except, perhaps, by a very direct reduction in demand). Here, our problem has never been with the technology of Artificial Neural Nets, which have been around for decades. It isn’t even with “AI” (note that no extant “AI” is actually “intelligent”)! No, our problem is with the tools. These tools are made with purpose and intent. Intent to defraud, intent to steal credit for the works of others, and the purpose of allowing corporations to save money on coding, staffing, and accountability for their actions, the purpose of having a black box a CEO can point to, shrug their shoulders, and say “what am I supposed to do? The AI agent told me to fire all of these people! Is it my fault that they were all <insert targetable group here>?!”
These tools cannot be used to know things. They are probabilistic models. These tools cannot be used to think for you. They are Chinese Rooms. For you to imply that the designers of these models are blameless — when their AI agents misidentify black men as criminals in facial recognition software; when their training data breaks every copyright law on the fucking planet, only to allow corporations to deepfake away any actual human talent in existence; when the language models spew vitriol and raging misinformation with the slightest accidental prompting, and can be hard-limited to only allow propagandized slop to be produced, or tailored to the whims of whatever despot directs the trolls today; when everyone now has to question whether they are even talking to a real person, or just a dim reflection, echoing and aping humanity like some unseen monster in the woods — is irreconcilable with even an iota of critical thought. Consider more carefully when next you speak, for your corporate-apologist principles will only help you long enough for someone to train your beloved “tool” on you. May you be replaced quickly.
You’ve made many incorrect assumptions and setup several strawmen fallacies. Rather than try to converse with someone who is only looking to feed their confirmation bias, I’ll suggest you continue your learnings by looking up the Dunning Kruger effect.
Can you point out and explain each strawman in detail? It sounds more like someone made good analogies that counter your point and you buzzword vomited in response.
Dissecting his wall of text would take longer than I’d like, but I would be happy to provide a few examples:
— Though wolfram claims to have read my post history, he seems to have completely missed my many posts hating on TSLA, robber barons, Reddit execs, etc. I completely agree with him that AI will be used for evil by corporate assholes, but I also believe it will be used for good (just like any other technology).
— Tools are neutral. They have more than one purpose. A nuclear bomb could be used to warm the atmosphere another planet to make it habitable. Not to mention any weapon can be used to defend humanity, or to attack it. Tools might be designed with a specific purpose in mind, but they can always be used for multiple purposes.
There are a ton of invalid assumptions about machine learning as well, but I’m not interested in wasting time on someone who believes they know everything.
I understand that you disagree with their points, but I’m more interested in where the strawman arguments are. I don’t see any, and I’d like to understand if I’m missing a clear fallacy due to my own biases or not.
Many of their points are factually incorrect. The first point I refuted is a strawman argument. They created a position I do not hold to make it easier to attack.
EDIT: now I understand. After going through your comments, I can see that you just claim confirmation bias rather than actually having to support your own arguments. Ironic that you seem to show all of this erudition in your comments, but as soon as anyone questions your beliefs, you just resort to logical buzzwords. The literal definition of the bias you claim to find. Tragic. Blocked.
Blocking individual on Lemmy is actually quite pointless as they still can reply to your comments and posts you just will not know about it while there can be whole pages long slander about you right under your nose
I’d say it’s by design to spread tankie propaganda unabated
You know what? They can go ahead and slander me. Fine. Good for them. They’ve shown they aren’t interested in actual argument. I agree with your point about the whole slander thing, and maybe there is some sad little invective, “full of sound and fury, signifying nothing”, further belittling my intelligence to try to console themself. If other people read it and think “yeah that dude’s right”, then that’s their prerogative. I’ve made my case, and it seems the best they can come up with is projection and baseless accusation by buzzword. I need no further proof of their disingenuity.
Least you ought to care about is who thinks what about your comments. I mean I don’t wanna tell you what to think but it seems sane to just ignore that
This forum means nothing, nothing we or anyone says here is of any importance. It’s not even something that is saved anywhere in the long run but is merely some speck of digital dust that will fly away as soon as you turn off your phone
It’s the most meaningless words you will ever write, here
So you know, you either enjoy drama or memes or some arcane knowledge or what is exactly you are doing on the internet dear stranger? I hope you aren’t thinking what you say here has any meaning because no one cares and I can assure you I won’t remember about any of this in 10 minutes tops
I feel like you are trying to say something serious here and I assure you this is inherently absurd. Whatever you wanna change or whatever you want to cope with start outside the phone/tablet/laptop
A take with which I disagree, but mostly a fair one. If you want to live an unexamined life, it’s not like that’s any of my business. You are absolutely correct on your read of me. I am committed to arcane esoterica, so that’s what I’m doing here. I’m here to discuss, and to learn, and I genuinely want to explore the limits of my understandings, identify my biases, and catch my fallacies, so when someone throws out some take that seems spectacularly inane, like “all technologies are tools, and all tools are the same”, I want to know if I’m missing something.
As far as why I dare to be so “absurd”: I’m inclined to wonder why you have decided that your personal nihilism means that no-one else should be able to have a serious discussion of any kind. That’s kind of sad.
You are on the internet how do you want to have serious discussions when you don’t even know who are you talking to. I may be a toddler for all you know it or some kind of psychotic serial killer chilling with a cigarette after the job well done. Like seriously people get the heck out of social media and do something actually that matters
How am I supposed talk seriously when you call my common sense a nihilism too lol. Just focus your capacity of seriousness for real life, thats the only sane way to go around. Or feel free to go insane on the internet discussing geopolitics with toddlers and ai and drunk college dropouts.
Enjoy getting baited every 5 seconds and probably be angry and miserable daily because you treat your phone with funny digital people so firkin seriously
Even this is getting too serious lol, like cmon I can’t believe I am actually doing this and every fiber of my being recoils right now saying “run, there’s nothing productive here, get the frick away and rain missiles from afar instead of getting involved in the pointless trenches of internet keyboard warfare”
Remind me to kill myself if I reply once more
Blocking means that you don’t have to devote your time and thoughts to that person. That’s pretty valuable. And even if they decide they are going to attack you, not-responding is often a good strategy vs that kind of crap anyway - to avoid getting pulled into an endless bad-faith argument. (I’d still suggest not announcing that you’ve blocked them though. Just block and forget about it.)
Maybe but it’s like putting your head in the sand here. You actually restrict yourself by doing it and not the other way around. Benefits are dubious
Every tech can be safe and unsafe? I think you’ve oversimplified to the point of meaninglessness. Obviously some technologies are safer than others, and some are more useful than others, and some have overwhelming negative effects. Different tech can and should be discussed and considered on a case by case basis - not just some “every tech is good and bad” nonsense.