the world is changing, and how some companies sourced there material may be unethical, but its the future. im not happy that two of my hobbys is getting replaced with a difrent format (coding and art) but its how its going to be. we need to learn how to use AI, instead of shunning its use
You can’t disregard ethics for progress, and also state that well since everyone else is doing it, it’s okay for me to do it too. Coding isn’t replaced by AI either, Chatgpt and other LLMs have proven to be terrible at programming.
Rollback Post to RevisionRollBack
— δ cyno • he/him • number one paladin fanδ — making a smoothie for meta ——————| EXTENDED SIG |—————— Φ • happily married to my lovely redpelt, minmaxer, microbiology student, and lover of anything colored red • Φ
I see two D&D account images and one Avatar: The Last Airbender one on the three posts before me. Huh.
says the guy with an AI generated pfp (Evidence: The tail starts to wrap around the rock, but immediately comes around on the exact same side it went in on, makes no sense. Also, why 3 separate small waterfalls? I don't really ever see that, especially at the PEAK of a mountain. I'm not a geography nerd, but why would a waterfall form at the peak of a mountain? Realistically lower down all the water would be able to gather. I mean, think about where waterfalls are in real life. The clouds are also conveniently forming specifically around the Dragons head, could be a stylistic choice, but doesn't make logical sense. Also has that AI like contrast)
(also slightly confused, they're just black with letters for you or they're just official D&D art or something?)
Okay, first thing I want to get out of the way: I wasn't judging anyone.
And yes, my image is AI generated. I have no more artistic talent than a cat dragging its claws through the mud randomly.
For me, there is a Halfling picture (Ash), a Dragonborn picture (Eris), and then the Avatar one (Matt).
Then get an actual artist. There are blind people who paint. Not being an artist is fine, but you can't use it as an excuse for AI. Go to my profile, back one page of posts, and there is a major discussion about AI, that probably discusses everything you'd bring up
...I already have the image, though. What's the point? Nothing more needs to be done. KEEPING the image won't do anything to anyone.
Debatable, but lets just leave this here, seeing as discussing it is not in anyone's interest but mine
I wasn't really discussing it anyway, so sure. Oh, and I want to make clear: I'm not upset about you bringing this up. Honestly, I get the same way when people act in a racist or sexist manner. Certain topics trigger certain people. Sorry if that came out degrading, I sometimes have trouble saying what I mean. English seems insufficient sometimes.
Its not degrading, dont worry
I think I sound unintentionally condescending online whenever I'm sort of debating or being passionate about something like that. I won't force my opinions on you, but reading some articles and listening to some experts on the subject might be a good idea. I'd suggest getting everyone's perspective, the companies, artists, non artists, and tech people all have different opinions and views to share. Just good to be educated on prevailing topics, yknow
there is ways to revert climate change (they are very large, and some people dont want to do them, but thats another problem), AI is something that we cant revert. I dont support AI art either, but attacking someone for having it as a profile picture is a step to far
He isn’t attacking you, and having it as a profile picture is kind of putting it out there that you use AI. Which is an objectively bad thing. And AI is even more fixable than climate change, with less steps. Boycotting is the first one - you shouldn’t be using it at all to showcase that you are against it. Plus why even use it when you can google “gold dragon on rock” and I’m sure thousands of images will come up.
Rollback Post to RevisionRollBack
— δ cyno • he/him • number one paladin fanδ — making a smoothie for meta ——————| EXTENDED SIG |—————— Φ • happily married to my lovely redpelt, minmaxer, microbiology student, and lover of anything colored red • Φ
The funny thing is I never said I supported it or didn't. My profile picture apparently started it, but I don't even really have an opinion on the matter.
Neutrality is an opinion and a stance.
Rollback Post to RevisionRollBack
— δ cyno • he/him • number one paladin fanδ — making a smoothie for meta ——————| EXTENDED SIG |—————— Φ • happily married to my lovely redpelt, minmaxer, microbiology student, and lover of anything colored red • Φ
I wasn't really discussing it anyway, so sure. Oh, and I want to make clear: I'm not upset about you bringing this up. Honestly, I get the same way when people act in a racist or sexist manner. Certain topics trigger certain people. Sorry if that came out degrading, I sometimes have trouble saying what I mean. English seems insufficient sometimes.
Its not degrading, dont worry
I think I sound unintentionally condescending online whenever I'm sort of debating or being passionate about something like that. I won't force my opinions on you, but reading some articles and listening to some experts on the subject might be a good idea. I'd suggest getting everyone's perspective, the companies, artists, non artists, and tech people all have different opinions and views to share. Just good to be educated on prevailing topics, yknow
Two things:
1. My family always says I end up sounding condescending towards them in the same situation, so I can sympathize.
2. I completely agree, to the point where I have two books on AI that I am planning on reading soon to get a better idea. (I bought them before this whole conversation).
Rollback Post to RevisionRollBack
Let the world shatter Into dust Nothing else matters Only us
The funny thing is I never said I supported it or didn't. My profile picture apparently started it, but I don't even really have an opinion on the matter.
Neutrality is an opinion and a stance.
I'm not neutral, though. I simply lack enough information to make a judgement. I try to NEVER participate in an argument without a reasonable understanding of the subject.
Rollback Post to RevisionRollBack
Let the world shatter Into dust Nothing else matters Only us
*looks at robot that has been partially codded by AI* about that....
Im not saying that what the AI gives is a finished product, we had to do some work on it, but it gets a lot done quickly. the same can be said for art, what the AI spits out needs work, and by using AI tools the entire process gets much faster.
(Keep in mind that I am against AI art, I just dont think we should mob anyone who uses it, its there choice*
Keyword: partially. AI cannot fully code things, and using AI to code stuff (as a programmer myself) is a good way to have faulty and clunky code. If people use AI in coding, it’s the same way people use google in coding. There is literally no difference. You shouldn’t sacrifice ethics for convenience either. Using AI tools for a “faster process” shouldn’t be your only consideration when you use it.
I think that we absolutely should call out and inform people that use AI so that they understand the ramifications. Every time you use a LLM, you are actively poisoning the environment. Everytime you use it, you are training a government tool that is being used to gather information and target people. I am against all AI in its current form because we have no guardrails and no precautions against it. People are suffering from AI usage and remaining “neutral” on it just means you’re willing to look past the long term problems for short term gain.
Rollback Post to RevisionRollBack
— δ cyno • he/him • number one paladin fanδ — making a smoothie for meta ——————| EXTENDED SIG |—————— Φ • happily married to my lovely redpelt, minmaxer, microbiology student, and lover of anything colored red • Φ
there is ways to revert climate change (they are very large, and some people dont want to do them, but thats another problem), AI is something that we cant revert. I dont support AI art either, but attacking someone for having it as a profile picture is a step to far
He isn’t attacking you, and having it as a profile picture is kind of putting it out there that you use AI. Which is an objectively bad thing. And AI is even more fixable than climate change, with less steps. Boycotting is the first one - you shouldn’t be using it at all to showcase that you are against it. Plus why even use it when you can google “gold dragon on rock” and I’m sure thousands of images will come up.
...I literally did get it from Googling, I just didn't check whether or not it was AI-Generated.
Rollback Post to RevisionRollBack
Let the world shatter Into dust Nothing else matters Only us
I would rather like to stop talking about this, as I have had freinds cut me off and refuse to speak to me for even talking about AI. all I see is a tool. you dont have to use the specific tool, but its a tool thats avalible and its people choices whether or not to use it
AI is not just a tool, like a calculator. It requires massive amounts of energy and resources and has a net negative output on society, regardless. I can link articles to support this if you’d like to read it.
Rollback Post to RevisionRollBack
— δ cyno • he/him • number one paladin fanδ — making a smoothie for meta ——————| EXTENDED SIG |—————— Φ • happily married to my lovely redpelt, minmaxer, microbiology student, and lover of anything colored red • Φ
The funny thing is I never said I supported it or didn't. My profile picture apparently started it, but I don't even really have an opinion on the matter.
Neutrality is an opinion and a stance.
I'm not neutral, though. I simply lack enough information to make a judgement. I try to NEVER participate in an argument without a reasonable understanding of the subject.
I can link some articles if you’d like? I’ll keep it to scientific journals and studies.
Rollback Post to RevisionRollBack
— δ cyno • he/him • number one paladin fanδ — making a smoothie for meta ——————| EXTENDED SIG |—————— Φ • happily married to my lovely redpelt, minmaxer, microbiology student, and lover of anything colored red • Φ
Keyword: partially. AI cannot fully code things, and using AI to code stuff (as a programmer myself) is a good way to have faulty and clunky code. If people use AI in coding, it’s the same way people use google in coding. There is literally no difference. You shouldn’t sacrifice ethics for convenience either. Using AI tools for a “faster process” shouldn’t be your only consideration when you use it.
I think that we absolutely should call out and inform people that use AI so that they understand the ramifications. Every time you use a LLM, you are actively poisoning the environment. Everytime you use it, you are training a government tool that is being used to gather information and target people. I am against all AI in its current form because we have no guardrails and no precautions against it. People are suffering from AI usage and remaining “neutral” on it just means you’re willing to look past the long term problems for short term gain.
I actually do a bit of coding, and absolutely refuse to use AI for that, as then I wouldn't learn anything. Although this may be besides the point.
Rollback Post to RevisionRollBack
Let the world shatter Into dust Nothing else matters Only us
The funny thing is I never said I supported it or didn't. My profile picture apparently started it, but I don't even really have an opinion on the matter.
Neutrality is an opinion and a stance.
I'm not neutral, though. I simply lack enough information to make a judgement. I try to NEVER participate in an argument without a reasonable understanding of the subject.
I can link some articles if you’d like? I’ll keep it to scientific journals and studies.
Actually, I already have two books on the subject, and a grandfather who is reasonably knowledgeable about it and willing to share information. As I don't really use AI, I think that will suffice. Thanks for the offer, though.
Rollback Post to RevisionRollBack
Let the world shatter Into dust Nothing else matters Only us
I would rather like to stop talking about this, as I have had freinds cut me off and refuse to speak to me for even talking about AI. all I see is a tool. you dont have to use the specific tool, but its a tool thats avalible and its people choices whether or not to use it
I respect that decision, and remember, you can back out and just not respond if there's already been statements to leave it be even if you get more responses (which you said twice)
I think you're expecting to find extremists in us simply for making this discussion, bringing your previous experiences unfairly into this one. I get that sometimes you disagree with someone and you just don't want to discuss it and stay friends, but some topics are worth more and are bigger than just being friends. I'm sure you'd understand If someone didn't just ignore a person being racist or homophobic so as not to stir up any tension
...I literally did get it from Googling, I just didn't check whether or not it was AI-Generated.
Oh I was under the impression you generated it yourself, my mistake.
Rollback Post to RevisionRollBack
— δ cyno • he/him • number one paladin fanδ — making a smoothie for meta ——————| EXTENDED SIG |—————— Φ • happily married to my lovely redpelt, minmaxer, microbiology student, and lover of anything colored red • Φ
I would rather like to stop talking about this, as I have had freinds cut me off and refuse to speak to me for even talking about AI. all I see is a tool. you dont have to use the specific tool, but its a tool thats avalible and its people choices whether or not to use it
I respect that decision, and remember, you can back out and just not respond if there's already been statements to leave it be even if you get more responses (which you said twice)
I think you're expecting to find extremists in us simply for making this discussion, bringing your previous experiences unfairly into this one. I get that sometimes you disagree with someone and you just don't want to discuss it and stay friends, but some topics are worth more and are bigger than just being friends. I'm sure you'd understand If someone didn't just ignore a person being racist or homophobic so as not to stir up any tension
Ngl being racist or homophobic is not as bad as using AI, unless you use it in really bad ways. My dad's an AI guy but I only talk about it with him when it's brought up and even then I try to be civil.
Rollback Post to RevisionRollBack
Hello! I am a perfectly sane gibberer. Hi! :D
Locations are dead, the Temple of Potassium has fallen but its ideals live on
I would rather like to stop talking about this, as I have had freinds cut me off and refuse to speak to me for even talking about AI. all I see is a tool. you dont have to use the specific tool, but its a tool thats avalible and its people choices whether or not to use it
I respect that decision, and remember, you can back out and just not respond if there's already been statements to leave it be even if you get more responses (which you said twice)
I think you're expecting to find extremists in us simply for making this discussion, bringing your previous experiences unfairly into this one. I get that sometimes you disagree with someone and you just don't want to discuss it and stay friends, but some topics are worth more and are bigger than just being friends. I'm sure you'd understand If someone didn't just ignore a person being racist or homophobic so as not to stir up any tension
Ngl being racist or homophobic is not as bad as using AI, unless you use it in really bad ways. My dad's an AI guy but I only talk about it with him when it's brought up and even then I try to be civil.
My dad is an AI guy too.
Rollback Post to RevisionRollBack
If I’m being annoying, tell me to shut up. Seriously. Just say “Bananer shut up.” And I will. For a few seconds!
Don’t listen to the folks down at Adohands. It’s good for me to overwork myself.
Professional idiot! Trans! Pansexual pancake! I am a minor so you will do none of that (GP) with me! I use He/They pronouns :3
Ngl being racist or homophobic is not as bad as using AI, unless you use it in really bad ways. My dad's an AI guy but I only talk about it with him when it's brought up and even then I try to be civil.
— δ cyno • he/him • number one paladin fanδ — making a smoothie for meta ——————| EXTENDED SIG |—————— Φ • happily married to my lovely redpelt, minmaxer, microbiology student, and lover of anything colored red • Φ
Ngl being racist or homophobic is not as bad as using AI, unless you use it in really bad ways. My dad's an AI guy but I only talk about it with him when it's brought up and even then I try to be civil.
*looks at robot that has been partially codded by AI* about that....
Im not saying that what the AI gives is a finished product, we had to do some work on it, but it gets a lot done quickly. the same can be said for art, what the AI spits out needs work, and by using AI tools the entire process gets much faster.
(Keep in mind that I am against AI art, I just dont think we should mob anyone who uses it, its there choice*
This post and your previous posts in this discussion suggest that you have bought into the spin that the various AI startups and such are selling, which is that their models are highly intelligent! They’re on the verge of true artificial intelligence! With just 1 billion dollars of seed money, we can make the jobreplacinator (tm) and prevent humans doing anything ever!
It’s also horseshit, to put it mildly. Apple’s R&D department recently put out a very interesting study/report, it’s worth a read. It proves fairly conclusively that the brand-name LLMS (Claude, Deepseek, etc), are not actually thinking at all; they’re instead ripping their solutions straight from the internet. Once they reach a certain threshold of difficulty in logic puzzles (one of the examples used in the study was Tower of Hanoi) they cannot complete it; no matter how much processing power is thrown at them. It’s another facet of “ChatGPT can’t make a full wine glass”.
At their core, LLMS are probability based compilation software. They scrape the internet (often illegally reusing copyrighted material, as has been previously covered - the only reason these companies aren’t backs against the wall in this regard is that the sheer quantity of amalgamation makes it very difficult to prove any one copyright infringement; second, those billions of dollars I mentioned) and spew out a response based on the probability of certain words following other words.
Both these models and the AI “industry” as a whole suffers from the age-old computing problem of shit in, shit out. While it may seem as if business is booming in the world of AI, it has actually added next to nothing to any company’s economic output; it’s only the sheer volume of money being poured into the concept that creates the illusion of success. It’s a very similar lead up to the dot com bubble, actually, hundreds (or even thousands) of startups and seed money cowboys taking advantage of the flood of capital without any real product involved. Add on behemoths like Google adding AI to every one of their ******* products as if it’s corn syrup, and you have a perfect economic illusion. Money in means money out, simply because those companies have dividends and wages to pay. WIRED magazine had an excellent issue on this, and they’re normally very gung-ho about new tech; their scepticism is worth mentioning.
As has been well publicised, LLMS suffer greatly from the “shit out” problem - much of the ‘innovation!’ that has gone into models like GPT-4 is much more likely guardrails to prevent the shit being so obvious - I imagine, not being able to stifle my conscience and scepticism enough to actually use the models for any purpose, that it is much less likely to make up citations now, for instance. Fundamentally, much like Sam Altman’s headline news apocalypse prepping a while ago, it’s a PR stunt to sell a product that doesn’t actually exist.
The problems it causes and ones it has drawn attention to, however - data farms, the environmental cost of chips, and more (Taiwan comes to mind), are very real, and by using it and defending it you reveal not only your own laziness but your misunderstanding of what the product is and what it does. To use your example, that code is not AI-original, it is stolen from many other people. Why would you get some shit product to make slop code when you can just make it right the first time without pointlessly using resources? Even proponents of the AI “industry” have said it would not function without ripping copyrighted material from all over the internet, and you’re enabling and encouraging them by using it.
It’s not the future, it’s dead shit. Dead shit that’s either a vehicle for money laundering or ticking, economic time bomb, depending on the extent of the tech industry’s deception on the concept’s capabilities.
You can’t disregard ethics for progress, and also state that well since everyone else is doing it, it’s okay for me to do it too. Coding isn’t replaced by AI either, Chatgpt and other LLMs have proven to be terrible at programming.
— δ cyno • he/him • number one paladin fan δ —
making a smoothie for meta
——————| EXTENDED SIG |——————
Φ • happily married to my lovely redpelt, minmaxer, microbiology student, and lover of anything colored red • Φ
Its not degrading, dont worry
I think I sound unintentionally condescending online whenever I'm sort of debating or being passionate about something like that.
I won't force my opinions on you, but reading some articles and listening to some experts on the subject might be a good idea. I'd suggest getting everyone's perspective, the companies, artists, non artists, and tech people all have different opinions and views to share. Just good to be educated on prevailing topics, yknow
forgot about a simple twist of fate
He isn’t attacking you, and having it as a profile picture is kind of putting it out there that you use AI. Which is an objectively bad thing. And AI is even more fixable than climate change, with less steps. Boycotting is the first one - you shouldn’t be using it at all to showcase that you are against it. Plus why even use it when you can google “gold dragon on rock” and I’m sure thousands of images will come up.
— δ cyno • he/him • number one paladin fan δ —
making a smoothie for meta
——————| EXTENDED SIG |——————
Φ • happily married to my lovely redpelt, minmaxer, microbiology student, and lover of anything colored red • Φ
Neutrality is an opinion and a stance.
— δ cyno • he/him • number one paladin fan δ —
making a smoothie for meta
——————| EXTENDED SIG |——————
Φ • happily married to my lovely redpelt, minmaxer, microbiology student, and lover of anything colored red • Φ
Two things:
1. My family always says I end up sounding condescending towards them in the same situation, so I can sympathize.
2. I completely agree, to the point where I have two books on AI that I am planning on reading soon to get a better idea. (I bought them before this whole conversation).
Let the world shatter
Into dust
Nothing else matters
Only us
Extended Signature: (^v^)
I'm not neutral, though. I simply lack enough information to make a judgement. I try to NEVER participate in an argument without a reasonable understanding of the subject.
Let the world shatter
Into dust
Nothing else matters
Only us
Extended Signature: (^v^)
Keyword: partially. AI cannot fully code things, and using AI to code stuff (as a programmer myself) is a good way to have faulty and clunky code. If people use AI in coding, it’s the same way people use google in coding. There is literally no difference. You shouldn’t sacrifice ethics for convenience either. Using AI tools for a “faster process” shouldn’t be your only consideration when you use it.
I think that we absolutely should call out and inform people that use AI so that they understand the ramifications. Every time you use a LLM, you are actively poisoning the environment. Everytime you use it, you are training a government tool that is being used to gather information and target people. I am against all AI in its current form because we have no guardrails and no precautions against it. People are suffering from AI usage and remaining “neutral” on it just means you’re willing to look past the long term problems for short term gain.
— δ cyno • he/him • number one paladin fan δ —
making a smoothie for meta
——————| EXTENDED SIG |——————
Φ • happily married to my lovely redpelt, minmaxer, microbiology student, and lover of anything colored red • Φ
...I literally did get it from Googling, I just didn't check whether or not it was AI-Generated.
Let the world shatter
Into dust
Nothing else matters
Only us
Extended Signature: (^v^)
AI is not just a tool, like a calculator. It requires massive amounts of energy and resources and has a net negative output on society, regardless. I can link articles to support this if you’d like to read it.
— δ cyno • he/him • number one paladin fan δ —
making a smoothie for meta
——————| EXTENDED SIG |——————
Φ • happily married to my lovely redpelt, minmaxer, microbiology student, and lover of anything colored red • Φ
I can link some articles if you’d like? I’ll keep it to scientific journals and studies.
— δ cyno • he/him • number one paladin fan δ —
making a smoothie for meta
——————| EXTENDED SIG |——————
Φ • happily married to my lovely redpelt, minmaxer, microbiology student, and lover of anything colored red • Φ
I actually do a bit of coding, and absolutely refuse to use AI for that, as then I wouldn't learn anything. Although this may be besides the point.
Let the world shatter
Into dust
Nothing else matters
Only us
Extended Signature: (^v^)
Actually, I already have two books on the subject, and a grandfather who is reasonably knowledgeable about it and willing to share information. As I don't really use AI, I think that will suffice. Thanks for the offer, though.
Let the world shatter
Into dust
Nothing else matters
Only us
Extended Signature: (^v^)
I respect that decision, and remember, you can back out and just not respond if there's already been statements to leave it be even if you get more responses (which you said twice)
I think you're expecting to find extremists in us simply for making this discussion, bringing your previous experiences unfairly into this one. I get that sometimes you disagree with someone and you just don't want to discuss it and stay friends, but some topics are worth more and are bigger than just being friends. I'm sure you'd understand If someone didn't just ignore a person being racist or homophobic so as not to stir up any tension
forgot about a simple twist of fate
Oh I was under the impression you generated it yourself, my mistake.
— δ cyno • he/him • number one paladin fan δ —
making a smoothie for meta
——————| EXTENDED SIG |——————
Φ • happily married to my lovely redpelt, minmaxer, microbiology student, and lover of anything colored red • Φ
It's fine. It's an easy possibility to overlook.
Let the world shatter
Into dust
Nothing else matters
Only us
Extended Signature: (^v^)
Ngl being racist or homophobic is not as bad as using AI, unless you use it in really bad ways. My dad's an AI guy but I only talk about it with him when it's brought up and even then I try to be civil.
Hello! I am a perfectly sane gibberer. Hi! :D
Locations are dead, the Temple of Potassium has fallen but its ideals live on
A mysterious link of chain... (Extended signature). PRAISE JEFF THE EVIL ROOMBA! REALLY cool video.
One of the Warlock Patrons on the forums. Low, low price of your soul, firstborn child and liver!
Titles: The Echoing Story Spewer (Drummer), the Endless Maws (Isis), the Mad Murderer (PJ), more on my extended sig
My dad is an AI guy too.
If I’m being annoying, tell me to shut up. Seriously. Just say “Bananer shut up.” And I will. For a few seconds!
Don’t listen to the folks down at Adohands. It’s good for me to overwork myself.
Professional idiot! Trans! Pansexual pancake! I am a minor so you will do none of that (GP) with me! I use He/They pronouns :3
Extended Signature!
Using AI kills people. Specifically minorities. https://www.techpolicy.press/progress-shouldnt-poison-black-communities/
— δ cyno • he/him • number one paladin fan δ —
making a smoothie for meta
——————| EXTENDED SIG |——————
Φ • happily married to my lovely redpelt, minmaxer, microbiology student, and lover of anything colored red • Φ
Oh (gp)
Hello! I am a perfectly sane gibberer. Hi! :D
Locations are dead, the Temple of Potassium has fallen but its ideals live on
A mysterious link of chain... (Extended signature). PRAISE JEFF THE EVIL ROOMBA! REALLY cool video.
One of the Warlock Patrons on the forums. Low, low price of your soul, firstborn child and liver!
Titles: The Echoing Story Spewer (Drummer), the Endless Maws (Isis), the Mad Murderer (PJ), more on my extended sig
This post and your previous posts in this discussion suggest that you have bought into the spin that the various AI startups and such are selling, which is that their models are highly intelligent! They’re on the verge of true artificial intelligence! With just 1 billion dollars of seed money, we can make the jobreplacinator (tm) and prevent humans doing anything ever!
It’s also horseshit, to put it mildly. Apple’s R&D department recently put out a very interesting study/report, it’s worth a read. It proves fairly conclusively that the brand-name LLMS (Claude, Deepseek, etc), are not actually thinking at all; they’re instead ripping their solutions straight from the internet. Once they reach a certain threshold of difficulty in logic puzzles (one of the examples used in the study was Tower of Hanoi) they cannot complete it; no matter how much processing power is thrown at them. It’s another facet of “ChatGPT can’t make a full wine glass”.
At their core, LLMS are probability based compilation software. They scrape the internet (often illegally reusing copyrighted material, as has been previously covered - the only reason these companies aren’t backs against the wall in this regard is that the sheer quantity of amalgamation makes it very difficult to prove any one copyright infringement; second, those billions of dollars I mentioned) and spew out a response based on the probability of certain words following other words.
Both these models and the AI “industry” as a whole suffers from the age-old computing problem of shit in, shit out. While it may seem as if business is booming in the world of AI, it has actually added next to nothing to any company’s economic output; it’s only the sheer volume of money being poured into the concept that creates the illusion of success. It’s a very similar lead up to the dot com bubble, actually, hundreds (or even thousands) of startups and seed money cowboys taking advantage of the flood of capital without any real product involved. Add on behemoths like Google adding AI to every one of their ******* products as if it’s corn syrup, and you have a perfect economic illusion. Money in means money out, simply because those companies have dividends and wages to pay. WIRED magazine had an excellent issue on this, and they’re normally very gung-ho about new tech; their scepticism is worth mentioning.
As has been well publicised, LLMS suffer greatly from the “shit out” problem - much of the ‘innovation!’ that has gone into models like GPT-4 is much more likely guardrails to prevent the shit being so obvious - I imagine, not being able to stifle my conscience and scepticism enough to actually use the models for any purpose, that it is much less likely to make up citations now, for instance. Fundamentally, much like Sam Altman’s headline news apocalypse prepping a while ago, it’s a PR stunt to sell a product that doesn’t actually exist.
The problems it causes and ones it has drawn attention to, however - data farms, the environmental cost of chips, and more (Taiwan comes to mind), are very real, and by using it and defending it you reveal not only your own laziness but your misunderstanding of what the product is and what it does. To use your example, that code is not AI-original, it is stolen from many other people. Why would you get some shit product to make slop code when you can just make it right the first time without pointlessly using resources? Even proponents of the AI “industry” have said it would not function without ripping copyrighted material from all over the internet, and you’re enabling and encouraging them by using it.
It’s not the future, it’s dead shit. Dead shit that’s either a vehicle for money laundering or ticking, economic time bomb, depending on the extent of the tech industry’s deception on the concept’s capabilities.
I can’t remember what’s supposed to go here.