Chatgpt is an infliction point not “it”
We come to bury ChatGPT, not to praise it
111–120 of 425 posts
Re: We come to bury ChatGPT, not to praise it
#112It's a tool that isn't going away. But even Yann Lecun has pointed out that "A house cat has way more common sense and understanding of the world than any LLM".
That's fundamentally the real issue. LLMs are far from even animal level "cognition" in their ability to solve novel problems. A lot of people seem to vastly underestimate the number of novel problems they solve on a regular basis.
In practice, what we're seeing now is something akin to the first wave of "ai hype" that led VCs into a spending frenzy years ago. Some of it did materialize. Most of it didn't.
LLMs and their related bretheren are pretty cool. They're very useful for solving routine problems. There's a lot of "routine" work out there in the world and it will probably replace some of those jobs.
But just like CNNs led to a frenzy of innovation in image recognition and LSTMs pushed NLP-related use cases forward this too will hit a wall.
If I were a betting man I think the main industries LLMs will disrupt are:
1. Search. ChatGPT is decent at summarizing stuff which will make searching more natural. 2. QA/Testing. A lot of this work is pretty repetitive and manual and LLM assistants can generate skeleton code for tests, etc. pretty well. 3. "basic" programming jobs that use "frameworks" to slap together simple apps/websites. A lot of this is repetitive gluing of stackoverflow code already so ChatGPT will make these folks considerably more productive. I'm not sure that these jobs will even disappear either.
I'm surprisingly less bullish on ChatGPT replacing marketing folks. It may be used to generate copy but the risks of putting out brand-damaging copy or a marketing campaign that doesn't make sense are too high.
Re: We come to bury ChatGPT, not to praise it
#113Earlier quoted context omitted.
> How functional does the code have to be before it's no longer a hallucination? From LLMs, at least, I expect it will always be a hallucination. Code is never the point. Code is the working medium by which people solve problems for other people. One way to see this is to realize that code bases on their own are generally worthless in the sense that people rarely pay much money for them. They pay for users. They pay…
> One way to see this is to realize that code bases on their own are generally worthless in the sense that people rarely pay much money for them. They pay for users. They pay for teams. But they don't generally pay for raw code This is something I've thought for a while. If Google's source code was leaked tomorrow, would it even matter? Almost certainly not, and most people probably couldn't do much with it either.
I’ve worked at another FAANG, and I used to joke all the time that if our code leaked it would give us a competitive advantage because the folks copying would have to spend so much time implementing our bloated code and god-awful build tools that they would probably go bankrupt.
Re: We come to bury ChatGPT, not to praise it
#114It has lots of value. But it's not the AI it pretends to be. Mostly because we and it can't always tell what's fact or fiction and it has 0 cognition. It's up to us to figure out where it can add value. It's way too early to announce its death. It's a super charged Cliff Clavin. https://en.wikipedia.org/wiki/Cliff_Clavin
> it has 0 cognition This seems to be the number one problem with lay understanding of ML masquerading as AGI. People immediately jump to the conclusion that the computer can think . Not yet, not even close. Edit: Switching AI to AGI to avoid any confusion.
That's not what AI means. You're talking about AGI.
AI is a field that covers all kinds of highly limited intelligences, including bad guys in a video game walking back and forth indefinitely.
Re: We come to bury ChatGPT, not to praise it
#115Here's the thing... so many jobs are bullshit. So many things people are paid to write, presentations made, etc. are bullshit. I'm not judging the bullshit per se. There's good bullshit and bad, necessary bullshit and not. But yeah, when you accept this reality, AI is sort of perfect for the bullshit that exists in our lives.
So bullshit, within a particular context, has value. And AI, within that context, is an infinite wealth generator.
Does this make that particular context (the world of bullshit) more valuable?
Is bullshit culture going to become more powerful now?
Re: We come to bury ChatGPT, not to praise it
#116Earlier quoted context omitted.
I tried to read the article too and found it both visually unpleasant and intellectually unfocused. GP is making the rather comically clever point that ChatGPT is useful to summarize text that might be difficult for humans to parse, even a text claiming that ChatGPT is a useless model only capable of spewing Islamophobic speech and in need of burying. I mean the article ends on some tangent about reactionary solution…
>both visually unpleasant Glad I wasn't the only one who noticed that. It's like the author picked a font/CSS combo to simulate a book with horrible typesetting.
Does it look like a blurred screenshot of text to anyone else?
Re: We come to bury ChatGPT, not to praise it
#117It's a tool that isn't going away. But even Yann Lecun has pointed out that "A house cat has way more common sense and understanding of the world than any LLM". That's fundamentally the real issue. LLMs are far from even animal level "cognition" in their ability to solve novel problems. A lot of people seem to vastly underestimate the number of novel problems they solve on a regular basis. In practice, what we're see…
It's pseudo-understanding, but pseudo-understanding can still be useful, as we see with, say, image classification.
Re: We come to bury ChatGPT, not to praise it
#118I found the article really difficult to read, so I summarised it with ChatGPT Large language models like the GPT family are statistical models that learn the structure of language by predicting missing words in sentences. These models are seen as "bullshit generators" as they have no idea what they are talking about and are designed to produce baseless assertions with confidence. The addition of reinforcement learnin…
I’m not sure why we need comments from people proud that they couldn’t be troubled to read the article that is supposedly being discussed, but it seems like a popular genre of reply whenever something tech-critical is posted.
While it may be frustrating to see comments from people who haven't read the article being discussed, it's important to recognize that everyone consumes and processes information in different ways. Some people may prefer to skim articles or jump straight to the comments section to get a sense of the wider conversation, rather than reading every detail of the article itself.
Some people may not have the time or the interest to read lengthy articles and may prefer to get the gist of the discussion through comments. It's important to keep an open mind and respect the ways in which others consume and engage with information.
—thus spake chatGPT
Re: We come to bury ChatGPT, not to praise it
#119I found the article really difficult to read, so I summarised it with ChatGPT Large language models like the GPT family are statistical models that learn the structure of language by predicting missing words in sentences. These models are seen as "bullshit generators" as they have no idea what they are talking about and are designed to produce baseless assertions with confidence. The addition of reinforcement learnin…
This is not even a correct summary of the content of the post. I'm genuinely entertained by the sheer amount of shamelessness demonstrated here. Seriously, though, technology at wrong hands endorses bad behaviors.
I went and wasted my time reading the post and it turns out the ChatGPT summary was accurate and this comment that it is not accurate was grossly wrong. Wish I hadn't wasted my time.
Re: We come to bury ChatGPT, not to praise it
#120I found the article really difficult to read, so I summarised it with ChatGPT Large language models like the GPT family are statistical models that learn the structure of language by predicting missing words in sentences. These models are seen as "bullshit generators" as they have no idea what they are talking about and are designed to produce baseless assertions with confidence. The addition of reinforcement learnin…
I’m not sure why we need comments from people proud that they couldn’t be troubled to read the article that is supposedly being discussed, but it seems like a popular genre of reply whenever something tech-critical is posted.
It's tech-person-splaining