The former is an opinion, the latter is absolutely true and you are utterly delusional if you think that gippity et al. literally do not do anything or even that they do not provide value. You are in a bubble, they can be (and are presently used in some cases) in a genuinely helpful way, even if they are misused by the vast majority and that many of the pitched use cases are complete nonsense used by corpos to inflate stock value.
No we can be against AI without it dividing the working class.
No you can’t, because pointing to AI as the problem instead of capitalism is undermining the discussions we should be having with pointless identity politics and culture wars over AI vs no-AI. It’s pointless fluff and detracts from the discussions we should be having about incentives and goals in and of society. And then we naturally wade in to the whole philosophical mess about the nature of art and discover that actually many people have entirely different conceptions of what art is, what isn’t art, what’s a craft and what isn’t and what it means to be an artist, and if there’s a difference between creativity and skill.
Thus, we stand divided, while they take more and more of everything.
This AI culture war, which exists only because the valuations of companies like OpenAI balloon every time it’s brought up as being the cause of all ills - this is because the cause of all ills is a powerful and valued thing and so the values go up, not to mention news about such things gets lots of clicks - and that’s why this dichotomy is being pushed on you, and that is why you think the things you do the way you do and why you use the words that you do.
This is the same process that ultimately dismantled left wing politics in the west after the fall of the soviet union - it worked by reframing the discussion around environmentalism and green politics, and now it’s the same - reframing the discussion to be about “technology” Vs. “no technology” when it’s obviously value neutral and only it’s uses and the policies that shape incentives that create those use cases that are the topic of discussion we should be having.
No, you should actually read my entire comment, and if you think “AI is part of capitalism” or that I think you are “not against capitalism” then read it again and keep reading it, and try to understand what I am saying rather than go off ‘vibes’ of what I am saying.
Think critically - think about it and process it, follow the logic from the base assumptions, the premises, to the end conclusion, and then if you pinpoint a logical flaw or imperical evidence that contradicts the base assumptions, then feel free to point it out.
AI is not part of that, “AI” is not capitalism - it is not part of capitalism, or part of communism or anything in between - it is a technology.
Think of it this way: are you against splitting the atom, or are you against doing so to kill millions of innocent civilians? I’d likely guess it’s the latter.
Try another: if a genuinely, earnestly socialist Soviet Union existed today and had their own LLMs or automation otherwise to assist with technological planning, would you be against socialism?
Technology is value neutral. It is how it’s used that’s the problem. Think critically.
You can of course be just anti-technology altogether, but that is irrelevant to discourse about capitalism.
I don’t disagree that technology is value neutral, I disagree that “AI” as such is even a technology or has any value.
Certainly software can have value, and that’s all this is. Time will tell if it has any practical application, but it has already revealed that it has significant negative externalities.
I don’t think we really even need to involve politics at all in the discussion if you’d prefer not to but since the fascist takeover of the USA is being enabled by the capital being generated through this AI bubble it seems relevant.
If an earnest socialist nation was developing LLMs I’d be critical of that too, and I would hope the workers would. Resource allocation vs return on this tech so far is diabolical.
It’s pretty crucial to note that the fascist takeover of the USA and even the AI bubble has happened after the “AI Spring”. Most of the investment that actually produced something was prior to well… most of the investment.
I agree with you on the fact that AI does seem like a highly inefficient way to do anything - but that only appears to be the case when we look at huge more recent models - just a few years ago FOSS locally run models were practically on par with their significantly bigger versions - models like the Mistral-7b for text or SDXL for image generation have lower hardware requirements than some video games these days and despite some shortcomings are still extremely impressive for what they are without sucking down an entire rainforest to summarize an email.
As for practical applications, they just make it possible to convert information into various formats, anything from translation to personalized tutorship for learning skills etc. and then there’s just the fact that they can make it possible for smaller software (and game) developement to take on much larger projects.
I’m not saying people should just accept slop either, I think there’s a goldilocks zone between this corpo idea of “employee-free” and “skill-free” and “code-free” software and “artist-free” art and complete and total rejection of using the tool at all.
Heck even just background removal in Photoshop is pretty neat even if it’s only marginally better than the magic eraser we’ve had for years.
Uh… No we can be against AI without it dividing the working class.
I’m at a loss really with this comment, it’s no worse than any other corpo slop? At least it does something? Neither of these statements are true
The former is an opinion, the latter is absolutely true and you are utterly delusional if you think that gippity et al. literally do not do anything or even that they do not provide value. You are in a bubble, they can be (and are presently used in some cases) in a genuinely helpful way, even if they are misused by the vast majority and that many of the pitched use cases are complete nonsense used by corpos to inflate stock value.
No you can’t, because pointing to AI as the problem instead of capitalism is undermining the discussions we should be having with pointless identity politics and culture wars over AI vs no-AI. It’s pointless fluff and detracts from the discussions we should be having about incentives and goals in and of society. And then we naturally wade in to the whole philosophical mess about the nature of art and discover that actually many people have entirely different conceptions of what art is, what isn’t art, what’s a craft and what isn’t and what it means to be an artist, and if there’s a difference between creativity and skill.
Thus, we stand divided, while they take more and more of everything.
This AI culture war, which exists only because the valuations of companies like OpenAI balloon every time it’s brought up as being the cause of all ills - this is because the cause of all ills is a powerful and valued thing and so the values go up, not to mention news about such things gets lots of clicks - and that’s why this dichotomy is being pushed on you, and that is why you think the things you do the way you do and why you use the words that you do.
This is the same process that ultimately dismantled left wing politics in the west after the fall of the soviet union - it worked by reframing the discussion around environmentalism and green politics, and now it’s the same - reframing the discussion to be about “technology” Vs. “no technology” when it’s obviously value neutral and only it’s uses and the policies that shape incentives that create those use cases that are the topic of discussion we should be having.
Do you somehow think I’m not against capitalism? AI is part of that
No, you should actually read my entire comment, and if you think “AI is part of capitalism” or that I think you are “not against capitalism” then read it again and keep reading it, and try to understand what I am saying rather than go off ‘vibes’ of what I am saying.
Think critically - think about it and process it, follow the logic from the base assumptions, the premises, to the end conclusion, and then if you pinpoint a logical flaw or imperical evidence that contradicts the base assumptions, then feel free to point it out.
AI is not part of that, “AI” is not capitalism - it is not part of capitalism, or part of communism or anything in between - it is a technology.
Think of it this way: are you against splitting the atom, or are you against doing so to kill millions of innocent civilians? I’d likely guess it’s the latter.
Try another: if a genuinely, earnestly socialist Soviet Union existed today and had their own LLMs or automation otherwise to assist with technological planning, would you be against socialism?
Technology is value neutral. It is how it’s used that’s the problem. Think critically.
You can of course be just anti-technology altogether, but that is irrelevant to discourse about capitalism.
I don’t disagree that technology is value neutral, I disagree that “AI” as such is even a technology or has any value.
Certainly software can have value, and that’s all this is. Time will tell if it has any practical application, but it has already revealed that it has significant negative externalities.
I don’t think we really even need to involve politics at all in the discussion if you’d prefer not to but since the fascist takeover of the USA is being enabled by the capital being generated through this AI bubble it seems relevant.
If an earnest socialist nation was developing LLMs I’d be critical of that too, and I would hope the workers would. Resource allocation vs return on this tech so far is diabolical.
It’s pretty crucial to note that the fascist takeover of the USA and even the AI bubble has happened after the “AI Spring”. Most of the investment that actually produced something was prior to well… most of the investment.
I agree with you on the fact that AI does seem like a highly inefficient way to do anything - but that only appears to be the case when we look at huge more recent models - just a few years ago FOSS locally run models were practically on par with their significantly bigger versions - models like the Mistral-7b for text or SDXL for image generation have lower hardware requirements than some video games these days and despite some shortcomings are still extremely impressive for what they are without sucking down an entire rainforest to summarize an email.
As for practical applications, they just make it possible to convert information into various formats, anything from translation to personalized tutorship for learning skills etc. and then there’s just the fact that they can make it possible for smaller software (and game) developement to take on much larger projects.
I’m not saying people should just accept slop either, I think there’s a goldilocks zone between this corpo idea of “employee-free” and “skill-free” and “code-free” software and “artist-free” art and complete and total rejection of using the tool at all.
Heck even just background removal in Photoshop is pretty neat even if it’s only marginally better than the magic eraser we’ve had for years.