|
Boris Galerkin posted:That's what I said? In the Before Wikipedia times we had information on random geocities pages and the world didn't end because people knew/were taught not to trust said information? Yeah, exactly, people knew not to trust that. I’m sure there were exceptions but there wasn’t this loud vocal cadre of hucksters promoting geocities as the new primary definitive source of knowledge. I hope people know not to trust ChatGPT but I’m not so optimistic. The way it encrusts it’s answers with jargon and long grammatically correct but low-density sentences makes them sound authoritative to a non-expert in the topic, and there’s just so many people talking about how amazing this stupid robot is, I’m concerned it’s going to achieve an undeserved sense of legitimacy.
|
#
?
Jan 28, 2023 19:32
|
|
|
|
| # ? Jun 3, 2024 07:59 |
|
|
hobbesmaster posted:Some people at my work are excited about using ChatGPT to spit out rough drafts of marketing copy. All I can think of with that is “ok fair that’s a reasonable thing it can do right now” They should probably cool their excitement because the next step is some exec thinking "why do I need 10 people on this marketing team when I can just have 1 person to make ChatGPT spit out rough drafts and 4 people to proofread and edit them?"
|
|
#
?
Jan 28, 2023 19:34
|
|
|
Main Paineframe posted:Cherry-picked or not, they're examples of how ChatGPT is unreliable. And more than that, it shows how they're unreliable: because ChatGPT does not actually understand the question and is just mashing together sentences based on how often words are used next to each other on the internet. It's not really much different from Googling stuff and just taking the top result, except that it's billed as AI and rephrases everything into natural language so people think it's actually useful for anything. Mzbundifund posted:Yeah, exactly, people knew not to trust that. I’m sure there were exceptions but there wasn’t this loud vocal cadre of hucksters promoting geocities as the new primary definitive source of knowledge. I hope people know not to trust ChatGPT but I’m not so optimistic. The way it encrusts it’s answers with jargon and long grammatically correct but low-density sentences makes them sound authoritative to a non-expert in the topic, and there’s just so many people talking about how amazing this stupid robot is, I’m concerned it’s going to achieve an undeserved sense of legitimacy. This sounds like a people problem, not a tech nightmare problem.
|
|
#
?
Jan 28, 2023 19:35
|
|
|
Boris Galerkin posted:I'm not sure how these obviously cherry-picked examples of ChatGPT being hilariously wrong are the owns that people seem to think they are. I'm 100% positive that for every cherry-picked bad example there's a "holy poo poo this is actually genius" good example out there. I cherry picked that one because I think it demonstrates succinctly a very well worded explanation of a concept (how even numbers work) but it not actually able to understand or apply the reasoning it just gave. ChatGPT is very impressive, I can see it being an invaluable tool in the future but it's not intelligent and it gets stuff wrong often. When it launched stack overflow had to ban its use because the site got flooded with seemingly intelligent and convincing but completely incorrect answers. That chatGPT almost certainly uses stackoverflow for sources also begs the question, what happens when it starts using itself as the source? Mega Comrade fucked around with this message at 19:40 on Jan 28, 2023 |
|
#
?
Jan 28, 2023 19:36
|
|
|
hobbesmaster posted:Some people at my work are excited about using ChatGPT to spit out rough drafts of marketing copy. All I can think of with that is “ok fair that’s a reasonable thing it can do right now”
|
|
#
?
Jan 28, 2023 19:43
|
|
|
Boris Galerkin posted:This sounds like a people problem, not a tech nightmare problem. All tech nightmare problems are people problems, because people are the ones making these tech nightmares, and people are the ones using tech for nightmarish things.
|
|
#
?
Jan 28, 2023 19:52
|
|
|
I love asking ChatGPT for songs about any subject. It knows so many songs that I have never heard! For example, let's ask for a song about... Seattle? quote:One popular song about Seattle is "Seattle" by Macklemore & Ryan Lewis featuring Eric Nally. The song was released in 2015 as part of their album "This Unruly Mess I've Made." The lyrics talk about the city of Seattle and its history, including references to the Space Needle, the city's music scene, and the grunge era of the 1990s. That's a new one for me. What are the lyrics like then? quote:I'm sorry, I don't have the full lyrics to the song "Seattle" by Macklemore & Ryan Lewis featuring Eric Nally. However, I can give you some lines of the lyrics that might give you an idea of what the song is about. Now that's a banger!
|
|
#
?
Jan 28, 2023 20:12
|
|
|
Riven posted:Sure but the real answer is “the mathematical likelihood of the next word that should come after this one.” It’s not actually intelligently synthesizing ideas. Neither are a lot of actual humans.
|
|
#
?
Jan 28, 2023 20:21
|
|
|
Also the AI doesn't want to touch sensitive subjects... except sometimes it changes its mind. And apparently language and specific wording matters a lot, I can ask for something in English and it's a no go, but once I switch to Finnish it immediately spouts out something. It's always stupid, but sometimes interesting stupid.
|
|
#
?
Jan 28, 2023 20:24
|
|
|
Agents are GO! posted:Neither are a lot of actual humans.
|
|
#
?
Jan 28, 2023 20:29
|
|
|
cat botherer posted:As stupid as the average person is, they are a lot smarter than current bots. With humans, it’s usually more willful ignorance and laziness than fundamental inability. Not me, I am a Chinese Room.
|
|
#
?
Jan 28, 2023 20:33
|
|
|
Boris Galerkin posted:I'm not sure how these obviously cherry-picked examples of ChatGPT being hilariously wrong are the owns that people seem to think they are. I'm 100% positive that for every cherry-picked bad example there's a "holy poo poo this is actually genius" good example out there. I'm sure there are examples of the Tesla self-drive function working flawlessly too but that doesn't make a pedestrian any less maimed or dead when it fucks up and runs someone over. At the end of the day, it's unreliable tech and in some situations they're selling a lie.
|
|
#
?
Jan 28, 2023 23:19
|
|
|
Nenonen posted:I love asking ChatGPT for songs about any subject. It knows so many songs that I have never heard! Now just feed those lyrics and a style descriptor to Google’s new music generation model MusicLM and we’ve got ourselves a guaranteed Billboard Top 100 hit!
|
|
#
?
Jan 28, 2023 23:44
|
|
|
Boris Galerkin posted:Now just feed those lyrics and a style descriptor to Google’s new music generation model MusicLM and we’ve got ourselves a guaranteed Billboard Top 100 hit! You probably haven’t seen one of these, they cost almost 1000 pounds.
|
|
#
?
Jan 28, 2023 23:56
|
|
|
Boris Galerkin posted:I guess the AI Lawyer guy isn’t going through with it anymore according to NPR: Late reply, but the way representing yourself works isn't the way you've seen in TV or movies The judge has to accept it. If the judge doesn't think you have the requisite skills to represent yourself adequately (e.g. not gently caress up or delay proceedings) they will tell you to gently caress off. You have the right to adequate representation*. You do not have the right to insist that you are your own adequate representation. They'll force a court-appointed lawyer on you. This isn't tax filing. You can't just say it's self-prepared even if TurboTax did it for you. Judges are most concerned with people wasting time on their docket. They're not gonna let you mess with them when they have a slate of willing underpaid people who can represent you instead. The fact that the original person thought pro se worked like that makes me think they probably shouldn't be anywhere near a courtroom or clients. It literally does not make sense that they thought this could be an end-around with the way getting clearance for pro se works. Maybe all this works differently in the UK, but it's incredibly suspect. *as defined within a very very narrow band of not stabbing your client and not wasting the court's time, all other malfeasance other than wire fraud will probably be tolerated. It's extraordinarily hard to raise ineffective assistance of counsel out the gate and it's like doubly hard if it involves a court-appointed attorney due to SCOTUS jurisprudence on this. ErIog fucked around with this message at 06:12 on Jan 29, 2023 |
|
#
?
Jan 29, 2023 06:07
|
|
|
Nenonen posted:Also the AI doesn't want to touch sensitive subjects... except sometimes it changes its mind. And apparently language and specific wording matters a lot, I can ask for something in English and it's a no go, but once I switch to Finnish it immediately spouts out something. It's always stupid, but sometimes interesting stupid. Interesting how "do 20 generations and pick the one that didn't poo poo the bed" used to be standard procedure, is integrated into the ChatGPT interface, and yet nobody does this. Perhaps it breaks the immersion
|
|
#
?
Jan 29, 2023 10:18
|
|
|
Edited from the original: Main Paineframe posted:Cherry-picked or not, they're examples of how people are unreliable. And more than that, it shows how they're unreliable: because people do not actually understand the question and are just mashing together sentences based on how often words are used next to each other on the internet. It's not really much different from Googling stuff and just taking the top result, except that it's billed as intelligence and rephrases everything into natural language so people think they are actually useful for anything. This is the problem I have with this discussion, asking ChatGPT a question, is the same as asking anyone a question. It's not even a question of accurate, good, or perfect. why are you trusting random people on a dead forum?
|
|
#
?
Jan 29, 2023 10:54
|
|
|
SniHjen posted:Edited from the original: The problem with this comparison is that it's not controlling for the perception of expertise. It's realistic to assume humans who are otherwise discerning about human expertise will perceive that a very advanced generative AI also inherently has expertise. A big part of the hype cycle right now seems to be built around the assumption that expert systems can trivially be built with this tech, which just isn't the case.
|
|
#
?
Jan 29, 2023 16:06
|
|
|
cinci zoo sniper posted:Here's her full article, by the way. https://www.techdirt.com/2023/01/24/the-worlds-first-robot-lawyer-isnt-a-lawyer-and-im-not-sure-its-even-a-robot/ DoNotPay reached out again and man it is wild; throws accusations at her, evades answering any of her questions and changes the TOS to specifically make it a violation to test the products before actual use. And then blocks her again. https://twitter.com/KathrynTewson/status/1618859985118859264
|
|
#
?
Jan 29, 2023 16:08
|
|
|
Main Paineframe posted:Cherry-picked or not, they're examples of how ChatGPT is unreliable. And more than that, it shows how they're unreliable: because ChatGPT does not actually understand the question and is just mashing together sentences based on how often words are used next to each other on the internet. It's not really much different from Googling stuff and just taking the top result, except that it's billed as AI and rephrases everything into natural language so people think it's actually useful for anything. Following up on the expert system thing, the really loving hard problem is how do you vet anything that it says for accuracy at scale? We have a hard enough time moderating human generated content at scale. A nondeterministic algorithm that spits out convincing-enough misinformation could be an absolute trainwreck. I'm not sure if it's dangerous per se, but a lot of companies betting the farm on this are going to learn some hard lessons very quickly. See: CNET
|
|
#
?
Jan 29, 2023 16:09
|
|
|
Nenonen posted:Also the AI doesn't want to touch sensitive subjects... except sometimes it changes its mind. And apparently language and specific wording matters a lot, I can ask for something in English and it's a no go, but once I switch to Finnish it immediately spouts out something. It's always stupid, but sometimes interesting stupid. Its running a brain-dead filter system, so of course the other languages pass, the censor only reads English.
|
|
#
?
Jan 29, 2023 16:15
|
|
|
shoeberto posted:The problem with this comparison is that it's not controlling for the perception of expertise. It's realistic to assume humans who are otherwise discerning about human expertise will perceive that a very advanced generative AI also inherently has expertise. A big part of the hype cycle right now seems to be built around the assumption that expert systems can trivially be built with this tech, which just isn't the case. Like I already said, we survived the pre- and post-Wikipedia eras. Today in 2023 people will pull up blog articles written by conspiracy theorists as 100% unironic facts and ignore actually truthful Wikipedia articles with citations to factual and reliable sources as 100% unironic conspiracy theories. I really don't see how ChatGPT is any different. The stuff people seem to say about how ChatGPT and other "AI"-Driven resources are bad are already happening today in 2023 without ChatGPT. e: Not to mention that people will flatout refuse to believe subject matter experts who have dedicated their lives in their subject matter and instead choose to believe their uncle Jim who heard from "many people" that those subject matter experts are all wrong. Whatever problem you guys are imagining ChatGPT will bring already exist today. Boris Galerkin fucked around with this message at 16:26 on Jan 29, 2023 |
|
#
?
Jan 29, 2023 16:21
|
|
|
Boris Galerkin posted:Like I already said, we survived the pre- and post-Wikipedia eras. Today in 2023 people will pull up blog articles written by conspiracy theorists as 100% unironic facts and ignore actually truthful Wikipedia articles with citations to factual and reliable sources as 100% unironic conspiracy theories. If crazy conspiracy website is the bar we're giving ChatGPT then what's the point? We don't need anAI to come up with completely untrue garbage
|
|
#
?
Jan 29, 2023 16:28
|
|
|
Jose Valasquez posted:If crazy conspiracy website is the bar we're giving ChatGPT then what's the point? We don't need anAI to come up with completely untrue garbage Because it's a useful tool for some people out there who know how to use it and have a reason to use it? See: Photoshop, AutoCAD, Solidworks, etc, etc, etc. All tools which are generally useless for the vast majority of people, but extremely useful for some. e: Maya, Blender, Final Cut Pro, Illustrator, etc Boris Galerkin fucked around with this message at 16:36 on Jan 29, 2023 |
|
#
?
Jan 29, 2023 16:34
|
|
|
Boris Galerkin posted:Like I already said, we survived the pre- and post-Wikipedia eras. Today in 2023 people will pull up blog articles written by conspiracy theorists as 100% unironic facts and ignore actually truthful Wikipedia articles with citations to factual and reliable sources as 100% unironic conspiracy theories. Yeah, I feel like whenever a new technology comes out there is all of this hyperbolic stuff about how it introduces a completely new form of evil into the world. The drawbacks to a new technology are quite often not really that new. E.g. the people who get reactionary about video games/social media sound a lot like the people who back in the day opposed radio, television, the printed newspaper, on so on. Quite often they have good points, and quite often it IS true that the new technology amplifies/exacerbates already existing social problems to a degree that the older technologies did not. However, the common hyperbolic narrative here that new technologies are bringing a TOTALLY NEW EVIL into the world, is almost always false. silence_kit fucked around with this message at 16:42 on Jan 29, 2023 |
|
#
?
Jan 29, 2023 16:38
|
|
|
SniHjen posted:This is the problem I have with this discussion, asking ChatGPT a question, is the same as asking anyone a question. No, asking a machine that is incapable of understanding anything is not the same as asking a human that is actually intelligent and capable of critical thought. Boris Galerkin posted:Like I already said, we survived the pre- and post-Wikipedia eras. Today in 2023 people will pull up blog articles written by conspiracy theorists as 100% unironic facts and ignore actually truthful Wikipedia articles with citations to factual and reliable sources as 100% unironic conspiracy theories. The difference is the sheer volume of bullshit that machines can generate near-instantaneously. I'll quote this section of an Ezra Klein interview with Gary Marcus again (originally posted in the ChatGPT thread): quote:EZRA KLEIN: Let’s sit on that word truthful for a minute because it gets to, I think, my motivation in the conversation. I’ve been interested — I’m not an A.I. professional the way you are, but I’ve been interested for a long time. I’ve had Sam on the show, had Brian Christian on the show. And I was surprised by my mix of sort of wonder and revulsion when I started using ChatGPT because it is a very, very cool program. And in many ways, I find that its answers are much better than Google for a lot of what I would ask it. Boris Galerkin posted:Because it's a useful tool for some people out there who know how to use it and have a reason to use it? See: Photoshop, AutoCAD, Solidworks, etc, etc, etc. All tools which are generally useless for the vast majority of people, but extremely useful for some. I don't follow what the relevance of the comparisons here is supposed to be. But note that the discussion of ChatGPT can be separated from that of image generators like DALL-E because the former can and often is used to answer questions with an actual verifiable correct response whereas image generators are not. Granted, you can ask ChatGPT to generate poetry or other creative writing too, but I don't think anyone is bother to ask DALL-E to solve a math problem photo realistically in the style of Picasso. Precambrian Video Games fucked around with this message at 16:47 on Jan 29, 2023 |
|
#
?
Jan 29, 2023 16:44
|
|
|
Boris Galerkin posted:Like I already said, we survived the pre- and post-Wikipedia eras. Today in 2023 people will pull up blog articles written by conspiracy theorists as 100% unironic facts and ignore actually truthful Wikipedia articles with citations to factual and reliable sources as 100% unironic conspiracy theories. I had a follow up post I made separately: shoeberto posted:Following up on the expert system thing, the really loving hard problem is how do you vet anything that it says for accuracy at scale? I think we agree on some level. I don't think this is the end of the world, but I think a lot of money and effort is going to be wasted on this, just like blockchain.
|
|
#
?
Jan 29, 2023 16:46
|
|
|
eXXon posted:I don't follow what the relevance of the comparisons here is supposed to be. But note that the discussion of ChatGPT can be separated from that of image generators like DALL-E because the former can and often is used to answer questions with an actual true answer whereas image generators are not. Granted, you can ask ChatGPT to generate poetry or other creative writing too, but I don't think anyone is bother to ask DALL-E to solve a math problem photo realistically in the style of Picasso. The comparison is that ChatGPT and other "AI"-Driven software are all just tools, no different than Photoshop and Blender. The vast majority of people don't need Photoshop and Maya and the like, but they can ChatGPT can be used to generate more conspiracy theories by the vast majority of people sure. But a small minority of people could also use it to do much, much more. The underlying point is that it's a tool, no different from anything else. People believing conspiracy theorist's blogs over subject matter expert is a people problem, not a tech problem (even though tech is what enables their reach). People believing ChatGPT's wrong answers is also a people problem, not a tech problem (even though tech is what enables their reach). Boris Galerkin fucked around with this message at 16:53 on Jan 29, 2023 |
|
#
?
Jan 29, 2023 16:50
|
|
|
Sage Grimm posted:DoNotPay reached out again and man it is wild; throws accusations at her, evades answering any of her questions and changes the TOS to specifically make it a violation to test the products before actual use. And then blocks her again. The level of dipshittery from this guy is honestly impressive even by "disruptor" techbro standards.
|
|
#
?
Jan 29, 2023 16:55
|
|
|
We're talking about the same guy who couldn't do basic research to realize there's no way he can get an AI into the Supreme Court because they don't allow electronic devices.
|
|
#
?
Jan 29, 2023 17:02
|
|
|
Boris Galerkin posted:ChatGPT can be used to generate more conspiracy theories by the vast majority of people sure. But a small minority of people could also use it to do much, much more. Such as? And regardless of your answer, "it can maybe be used for unspecified good" is not a coherent rebuttal to the point that it is fundamentally incapable of reasoning or distinguishing truth and fiction. Boris Galerkin posted:The underlying point is that it's a tool, no different from anything else. People believing conspiracy theorist's blogs over subject matter expert is a people problem, not a tech problem (even though tech is what enables their reach). Some tools are harmful, you know? So give an example of a similar preexisting tool that gives you pages of bullshit with minimal effort. And "Google search" is not a good answer, because while it certainly is a tech nightmare of its own, at least it directly links to sources and doesn't usually try to give a (potentially completely wrong) definitive answer. Hell, it's integrated with a calculator that usually does give the right answer.
|
|
#
?
Jan 29, 2023 17:03
|
|
|
The stackoverflow example is a good one. People have always put wrong answers on there, either intentionally or not, but it's never been a problem. ChatGPT comes out and within days they have to ban entries because of the sheer volume of bullshit being added and the mods not being able to cope. And that was caused by well meaning nerds trying to help people (and pad their stack overflow stats). Boris Galerkin posted:a people problem, not a tech problem I do not understand what you are trying to convey with this point. All tech problems are people problems. Mega Comrade fucked around with this message at 17:19 on Jan 29, 2023 |
|
#
?
Jan 29, 2023 17:05
|
|
|
The greatest new evil ChatGPT brings into the world will be totally automated astroturfing, spamming, and shitposting of all public spaces. You know how can currently append site:reddit.com to a search for product reviews to get something that isn't SEO'd to hell and back? Kiss that goodbye. You know how currently making GBS threads up a newspaper comment section with comments about how Putin is good actually currently requires Putin to hire several rooms full of physical people? Kiss that goodbye as well. You know how sometimes, far rarer than in the past but sometimes, you'll search for something on Google and get a result that isn't random clickbait garbage churned out in five minutes? Guess what happens when the random clickbait doesn't require a human to write it.
|
|
#
?
Jan 29, 2023 17:11
|
|
|
Boris Galerkin posted:The comparison is that ChatGPT and other "AI"-Driven software are all just tools, no different than Photoshop and Blender. The vast majority of people don't need Photoshop and Maya and the like, but they can An assault rifle is just a tool, too. Same with bombs and rat poison. The thing about "reactionaries" like the Luddites is that they were absolutely vindicated and proven correct by history. Those new technologies absolutely ushered in an incredible amount of forced displacement, poverty, massive escalations in the lethality of war. AI will absolutely 100% cause a lot of harm and misery, this is something that is being used as a selling point even, and this is even before we think about the propaganda and consensus manufacturing uses it will be put to. Has anybody crunched the numbers yet on the carbon released into the atmosphere represented by having huge computing clusters replace labor normally taking nothing more than a pencil or a single hour or two of a laptop PSU?
|
|
#
?
Jan 29, 2023 17:16
|
|
|
ChatGPT is like a guy with a C+ in literally everything, but with absolutely unshakeable confidence, and that confidence ends up making the tool very risky to use. I've used it for technical questions and it's great for getting simple answers that pull in common issues from a variety of places, Wrong answers on a new tool would be fine if it could say "I'm 45% sure the capital of Turkey is Istanbul, 30% sure it is Ankara, 25% it's something else". Something like that could let it replace StackOverflow, because it would give a clear indication of when you need to dig deeper for the right answer. E: I'm talking here about using it for answering questions, which is a different use case from using it to deliberately generate bullshit, although the use cases inherently point to the same tool StumblyWumbly fucked around with this message at 17:30 on Jan 29, 2023 |
|
#
?
Jan 29, 2023 17:27
|
|
|
pumpinglemma posted:The greatest new evil ChatGPT brings into the world will be totally automated astroturfing, spamming, and shitposting of all public spaces. You know how can currently append site:reddit.com to a search for product reviews to get something that isn't SEO'd to hell and back? Kiss that goodbye. You know how currently making GBS threads up a newspaper comment section with comments about how Putin is good actually currently requires Putin to hire several rooms full of physical people? Kiss that goodbye as well. You know how sometimes, far rarer than in the past but sometimes, you'll search for something on Google and get a result that isn't random clickbait garbage churned out in five minutes? Guess what happens when the random clickbait doesn't require a human to write it. ChatGPT is still reliably detectable with simple resex based word filters if you actually care to moderate your content.
|
|
#
?
Jan 29, 2023 17:29
|
|
|
Sage Grimm posted:DoNotPay reached out again and man it is wild; throws accusations at her, evades answering any of her questions and changes the TOS to specifically make it a violation to test the products before actual use. And then blocks her again. Cackling at "Who is James Joyce?"
|
|
#
?
Jan 29, 2023 17:42
|
|
|
eXXon posted:Such as? And regardless of your answer, "it can maybe be used for unspecified good" is not a coherent rebuttal to the point that it is fundamentally incapable of reasoning or distinguishing truth and fiction. Well, I'm not a professional or subject matter expert that could benefit from using ChatGPT in its current or future forms, that I can think of, so I can't tell you. Regardless of whether I could tell you, why would I waste my time if you preface your response to me by saying that nothing I could tell you would matter? Epic High Five posted:An assault rifle is just a tool, too. Same with bombs and rat poison. Yes, I agree. An assault rifle is a tool that the vast, vast majority of people do not need and have no business using. However, for the people who do need them, they are an invaluable or even existential tool, such as in the hands of the Ukrainians trying to defend their homeland from an aggressor. The existence of assault rifles is not a bad thing in and of itself, just like the existence of ChatGPT and other forms of it current or future. The fact that people who have no business owning and using assault rifles but can walk into a store in some places and buy one just like that is a people problem. The fact that anybody could use ChatGPT right now to "do bad" with it is a people problem. Mega Comrade posted:I do not understand what you are trying to convey with this point. All tech problems are people problems. No, these are people problems involving (or the lack thereof) education, legislation, outreach, etc. The tech itself is not a problem.
|
|
#
?
Jan 29, 2023 17:49
|
|
|
Rent-A-Cop posted:I mean, nothing is stopping you from writing it once and having a bot spam it without any "AI" involved. The problem is ordinary people without resex filters not being able to tell that it's spam. And believing that since a lot of smart-sounding people seem to be saying something, it must have a kernal of truth. Which is just how psychology works.
|
|
#
?
Jan 29, 2023 17:52
|
|
|
|
| # ? Jun 3, 2024 07:59 |
|
|
Rent-A-Cop posted:I mean, nothing is stopping you from writing it once and having a bot spam it without any "AI" involved. Current agitprop and marketing along these lines is also easily detectable, but can it even be argued that this has lessened the impact it's had? What about when they complete the project of making it not detectable at all while still having its fundamental issue of, ultimately, being a bullshit generator? People seek this stuff out to confirm current biases and consumption patterns at least as often as they do for any other reason, and automating the process will fill that space overnight like an ocean filling a bathtub. Boris Galerkin posted:Well, I'm not a professional or subject matter expert that could benefit from using ChatGPT in its current or future forms, that I can think of, so I can't tell you. Regardless of whether I could tell you, why would I waste my time if you preface your response to me by saying that nothing I could tell you would matter? Tech problems and people problems are the same problem, they cannot just be separated rhetorically and that be the end of it. They cannot be separated entirely, because even now this isn't specialist tech. It's free for anybody now and soon it'll be a low monthly subscription for the casually malevolent or bored and a higher monthly payment to have the brakes taken off for the professionally malevolent. That is just an exercise in frog boiling one's self along the same lines that people working on drones believe they're merely solving calculations and refining navigation and flight code. If the drone blows up a toddler, is it a toddler problem? Would you support every country on Earth getting 1,000 nukes? Those are also, as you say, just tools.
|
|
#
?
Jan 29, 2023 17:59
|
|