4 ms·

>This week, OpenAI announced that they have created a public facing API for ChatGPT. At this point, I think it's over. We are going to have to learn to live wit

>This week, OpenAI announced that they have created a public facing API for ChatGPT. At this point, I think it's over. We are going to have to learn to live with large language models and all of the other types of models that people will inevitably cook up. What were they hoping for exactly? That OpenAI would just close down and stop serving requests?

favaq 4y ago
His explanations of why this is so, so bad fall very short. Why close down a tool that has good and bad uses? That would be like abolishing knives.
- pmoriarty 4y ago
  There's a more thoughtful analysis of reasons to oppose chatgpt here: [1], which concludes: "I'm not a fan of banning promising technologies that also have very positive use-cases. Especially when I could out-source really boring tasks to it as well. But the enormous amount of potential negative impact does worry me a lot so that a short-term ban is the lesser evil here to me." [1] - https://karl-voit.at/2023/01/14/chatgpt/ https://karl-voit.at/2023/01/14/chatgpt/
  - throwaway4aday 4y ago
    > Furthermore, even experts get this check-task wrong simply because humans tend to assume the correct answer, overlooking hidden mistakes too easily. Everybody has made the experience that we are unable to find certain typing mistakes where it is much easier to find typos when reading text of other people. > Another issue at hand is that so far, text written by people who lack a certain level of knowledge was easy to spot and detect. Those texts typically had typing errors, were using less elaborate language and followed a certain pattern to be recognized as bullshit. This seems to be the core of the author's argument. It's not very convincing. The rest of the article is just various forms of paranoia and hand wringing about supposed bad actors or incompetence. This is simple historical blindness, there have always been bad actors and incompetent people and yes any technological advance extends their ability to do bad things but those advances also extend the ability of others to counteract those effects. We've already had the ultimate "bad technology" for almost 80 years and it's currently in the hands of some bad actors. Lots of people were concerned about it leading to the end of the world, and still are, but almost no one predicted that it would result in one of the most peaceful periods in human history due to the horrendous consequences of attempting to wage an all out war. Regardless, there's no putting the genie back in the bottle once it's out and the LLM genie is most definitely out
joshka 4y ago
Side note it’s xe or they or she rather than he. See https://xeiaso.net/pronouns https://xeiaso.net/pronouns
xena 4y ago
Honestly I was hoping that ChatGPT was just a fad. It exploded in users almost instantly (2 months to one million MAU, that's _insane_), and I was honestly hoping that interest would wane naturally over the next few months. I thought that this was part of the GPT-4 training process or something. That clearly isn't going to happen. My other hope was that these large language models would lose popularity in general and everyone would move on, but now I'm seeing LLM output in screenshots in slack threads when I ask a clarifying fundamental question about CSS. Like I said in the article: Pandora's box is now open and we get to learn to live with the diseases. Side note: it would be nice if you edited your comment to refer to me accurately with "they" or "she". I'd really appreciate it!
- sebzim4500 4y ago
  >Side note: it would be nice if you edited your comment to refer to me accurately with "they" or "she". I'd really appreciate it! I would if I could, but I don't see an 'edit' button.
  - Kye 4y ago
    There's a two hour edit window on comments.
  - dang 4y ago
    I'm terribly sorry for getting to this so late, but I've edited it now.

favaq 4y ago

His explanations of why this is so, so bad fall very short. Why close down a tool that has good and bad uses? That would be like abolishing knives.

pmoriarty 4y ago
There's a more thoughtful analysis of reasons to oppose chatgpt here: [1], which concludes: "I'm not a fan of banning promising technologies that also have very positive use-cases. Especially when I could out-source really boring tasks to it as well. But the enormous amount of potential negative impact does worry me a lot so that a short-term ban is the lesser evil here to me." [1] - https://karl-voit.at/2023/01/14/chatgpt/ https://karl-voit.at/2023/01/14/chatgpt/
- throwaway4aday 4y ago
  > Furthermore, even experts get this check-task wrong simply because humans tend to assume the correct answer, overlooking hidden mistakes too easily. Everybody has made the experience that we are unable to find certain typing mistakes where it is much easier to find typos when reading text of other people. > Another issue at hand is that so far, text written by people who lack a certain level of knowledge was easy to spot and detect. Those texts typically had typing errors, were using less elaborate language and followed a certain pattern to be recognized as bullshit. This seems to be the core of the author's argument. It's not very convincing. The rest of the article is just various forms of paranoia and hand wringing about supposed bad actors or incompetence. This is simple historical blindness, there have always been bad actors and incompetent people and yes any technological advance extends their ability to do bad things but those advances also extend the ability of others to counteract those effects. We've already had the ultimate "bad technology" for almost 80 years and it's currently in the hands of some bad actors. Lots of people were concerned about it leading to the end of the world, and still are, but almost no one predicted that it would result in one of the most peaceful periods in human history due to the horrendous consequences of attempting to wage an all out war. Regardless, there's no putting the genie back in the bottle once it's out and the LLM genie is most definitely out

joshka 4y ago

Side note it’s xe or they or she rather than he. See https://xeiaso.net/pronouns https://xeiaso.net/pronouns

xena 4y ago

Honestly I was hoping that ChatGPT was just a fad. It exploded in users almost instantly (2 months to one million MAU, that's _insane_), and I was honestly hoping that interest would wane naturally over the next few months. I thought that this was part of the GPT-4 training process or something. That clearly isn't going to happen. My other hope was that these large language models would lose popularity in general and everyone would move on, but now I'm seeing LLM output in screenshots in slack threads when I ask a clarifying fundamental question about CSS. Like I said in the article: Pandora's box is now open and we get to learn to live with the diseases. Side note: it would be nice if you edited your comment to refer to me accurately with "they" or "she". I'd really appreciate it!

sebzim4500 4y ago
>Side note: it would be nice if you edited your comment to refer to me accurately with "they" or "she". I'd really appreciate it! I would if I could, but I don't see an 'edit' button.
- Kye 4y ago
  There's a two hour edit window on comments.
- dang 4y ago
  I'm terribly sorry for getting to this so late, but I've edited it now.