'We support the work of those who are trying to limit the harm generative AI is doing to videogames and our world, and apologise that Big Walk’s LLM usage may not meet their standards'
this sounds like they used it as a search engine rather than as an assistant. it’s one of those weird gray areas that come up with how integrated it is now. like, if you try to find something with a search engine and you get the “ai result” thing, is that “ai use”? the tokens were used anyway, and you read it even if you then click a link.
i hope the otherwise great reception of big walk can help a bit with the debate here.
Sometimes it’s so funny to see you all type “well it’s a grey area” and then stretch it this way and that. But ofc, Gregory, all other areas of life are black or white, are they? Or just the ones you don’t understand?
Just here me out. What if all the whine and opinions and explanations all point toward a grey area because the entire thing is grey, because it’s a technology we “found” rather than created. There’s a reason most machine learning research is trying to understand what the actual fuck the thing is doing.
Yeah, it’s a search engine sometimes to you. Technically all image generations are too. They just search color patterns instead. If you make it talk to itself, suddenly it becomes a programmer. It’s a field that is just completely not possible to draw any borders or lines across, because we don’t understand what the fuck it’s doing
Everyone keeps trying to mansplain the process or the architecture, or algorithms involved, as if that explains it to them. But they are mistaken. It eludes all of humanity still. We have no clue whatsoever what makes it “search” (in your view) or “infer” or “predict” or “model” answers if we set it up to do so. If you can find out, i urge you to reach out to the researchers. We could all sleep a lot better if we could understand it and prevent sociopathic psychotic behaviors.
Coding with agents if you were a programmer before is a harrowing but also empowering, it’s just impossible to not use the ability to slow through boring parts of the work or scan massive amounts of data. It is just useful as fuck if you already knew what you were doing. If you didn’t, it’s just dangerous. But using it to code is not even close to using it for writing or other art. And that shit is the exact same technology.
Tldr: We don’t understand it. It’s no wonder we can’t say where the line goes because of that.
This is the infantile mansplaining-like phenomenon I was eluding to. You don’t know. You think that it is a compression algorithm. Another, that it is a stochastic parrot and yet another that it is a live being. It is rare except for the people that actually read and understand the code that people admit, that the inner workings of the neural net is still a complete mystery. Oh, we know things like that the k-space seem to emphasize phrases that are adjacent to but not necessarily part of output. We know that we can afford certain sidebands for orthogonal functions we also don’t understand, except that they improve this or that tendency. It is extremely not a compression algorithm for several reasons. The first one being the inability to reform the original message with confidence, or with the slightest hint of efficiency. We do not know how it works, and it is easy to spot emotional counterfactual discourse by the confidence with which they proclaim to understand how modern machine learning models work.
i feel like there is a fundamental difference between “we don’t know why it chooses the coefficients it does” and “we don’t know how it works”. if anything, that’s what’s infantile. that, and accusing people of it. the fact that gzip can be used as a language model should be a pretty big indicator.
Everyone keeps trying to mansplain the process or the architecture, or algorithms involved
Because you’re making assumptions about how it works and those assumptions are wrong. I’m sorry if you find that offensive but you need to be educated before you can make an informed decision you seem to be unhappy with this.
Coding with agents if you were a programmer before is a harrowing but also empowering
So you do know how to code, which makes your prior incredulousness of the process kind of hard to comprehend.
No, I am telling you that my understanding and that of the researchers is that we have a black box we are trying to understand. We mimiced nature and found something but don’t fully know what is going on inside. Anyone that says they do, are lying. You cannot look at “the code” for the transformer or any process and understand why the weights produce text that solves problems. It is idiotic to pretend you understand without understanding how the technology functions at all
this sounds like they used it as a search engine rather than as an assistant. it’s one of those weird gray areas that come up with how integrated it is now. like, if you try to find something with a search engine and you get the “ai result” thing, is that “ai use”? the tokens were used anyway, and you read it even if you then click a link.
i hope the otherwise great reception of big walk can help a bit with the debate here.
Sometimes it’s so funny to see you all type “well it’s a grey area” and then stretch it this way and that. But ofc, Gregory, all other areas of life are black or white, are they? Or just the ones you don’t understand?
Just here me out. What if all the whine and opinions and explanations all point toward a grey area because the entire thing is grey, because it’s a technology we “found” rather than created. There’s a reason most machine learning research is trying to understand what the actual fuck the thing is doing.
Yeah, it’s a search engine sometimes to you. Technically all image generations are too. They just search color patterns instead. If you make it talk to itself, suddenly it becomes a programmer. It’s a field that is just completely not possible to draw any borders or lines across, because we don’t understand what the fuck it’s doing
Everyone keeps trying to mansplain the process or the architecture, or algorithms involved, as if that explains it to them. But they are mistaken. It eludes all of humanity still. We have no clue whatsoever what makes it “search” (in your view) or “infer” or “predict” or “model” answers if we set it up to do so. If you can find out, i urge you to reach out to the researchers. We could all sleep a lot better if we could understand it and prevent sociopathic psychotic behaviors.
Coding with agents if you were a programmer before is a harrowing but also empowering, it’s just impossible to not use the ability to slow through boring parts of the work or scan massive amounts of data. It is just useful as fuck if you already knew what you were doing. If you didn’t, it’s just dangerous. But using it to code is not even close to using it for writing or other art. And that shit is the exact same technology.
Tldr: We don’t understand it. It’s no wonder we can’t say where the line goes because of that.
it’s a compression algorithm, my guy. we don’t know why 7z picks the numbers it does for its tables either, but we damn well know how.
No, it’s absolutely not a compression algorithm.
it certainly works on the same principles.
This is the infantile mansplaining-like phenomenon I was eluding to. You don’t know. You think that it is a compression algorithm. Another, that it is a stochastic parrot and yet another that it is a live being. It is rare except for the people that actually read and understand the code that people admit, that the inner workings of the neural net is still a complete mystery. Oh, we know things like that the k-space seem to emphasize phrases that are adjacent to but not necessarily part of output. We know that we can afford certain sidebands for orthogonal functions we also don’t understand, except that they improve this or that tendency. It is extremely not a compression algorithm for several reasons. The first one being the inability to reform the original message with confidence, or with the slightest hint of efficiency. We do not know how it works, and it is easy to spot emotional counterfactual discourse by the confidence with which they proclaim to understand how modern machine learning models work.
i feel like there is a fundamental difference between “we don’t know why it chooses the coefficients it does” and “we don’t know how it works”. if anything, that’s what’s infantile. that, and accusing people of it. the fact that gzip can be used as a language model should be a pretty big indicator.
This is riddled with logical fallacies and none of these arguments hold. You should be ashamed.
No thanks
Because you’re making assumptions about how it works and those assumptions are wrong. I’m sorry if you find that offensive but you need to be educated before you can make an informed decision you seem to be unhappy with this.
So you do know how to code, which makes your prior incredulousness of the process kind of hard to comprehend.
No, I am telling you that my understanding and that of the researchers is that we have a black box we are trying to understand. We mimiced nature and found something but don’t fully know what is going on inside. Anyone that says they do, are lying. You cannot look at “the code” for the transformer or any process and understand why the weights produce text that solves problems. It is idiotic to pretend you understand without understanding how the technology functions at all