3 ms·

It's weird to me that you are conflating asking GPT a question related to your article and writing your article. Would you have a similar reaction if he had sai

It's weird to me that you are conflating asking GPT a question related to your article and writing your article. Would you have a similar reaction if he had said "according to google"? There does not seem to be any evidence at all that the author didn't write this entire article, and the fact that they explicitly reference that they consulted GPT on some related point seems further evidence that they _didn't_ have GPT write it (I think if they _had_ had GPT write it, they would have avoided mentioning GPT at all)

howenterprisey 3y ago
Not them, but no, I wouldn't have a similar reaction if it were "according to google", because in the context of a blog posted to HN I'd expect it to mean a cursory bit of research, which is way better.
- MostlyStable 3y ago
  In the very near future (<5 years in my opinion), nearly all search requests are going to be mediated by a GPT like AI agent. It's already true that Bing search is better for some kinds of searches than a standard google search (which has become increasingly shit over the past few years as Google has lost the SEO war to blogspam). But even if we aren't quite there yet, referencing that you asked a related question to GPT in a an article you have written is still not fundamentally different than doing some research on google. GPT will still give bad answers to certain kinds of questions (and while the author didn't say what they asked or what exactly GPT responded, my best guess is that this was a pretty useless thing to ask GPT), but you can also find some real stupid shit on Google. That sentence was almost more rhetorical flourish than anything else, and while it didn't land particularly well for me personally, I find it _far_ weirder that the commenter I replied to went off about authorship and trust.
  - howenterprisey 3y ago
    GPT and Google results are extremely fundamentally different. I can use various heuristics to evaluate how trustworthy a site I find on Google is, and if I need to I can do any amount of research on the site itself, who wrote it, etc (and if I can't, then that's useful information too). In contrast, GPT has no indication of where it got anything from, but even that's giving too much credit. It's just stringing words together. Whether it's wrong or right, I have no way of knowing. If I'm reading a website and I know the first thing it says is for sure correct (or incorrect) then I can judge the rest of it appropriately. That's absolutely not the case for GPT, because the correctness of each sentence is relatively (entirely?) statistically independent. That's only the beginning of it. There are so many more ways that doing actual research is superior; there's just no contest. I will be horrified if most searches are mediated by LLMs in the future and I will do my best to never use such a search product. (I use Kagi myself and haven't had a problem yet.)
    - MostlyStable 3y ago
      Bing provides links to sources so you can evaluate the things yourself. I expect this to be the way most search will work. The AI will mostly just be filtering the searches you use. Which google already does, the AI will just do it better, since, as I mentioned earlier, Google is getting increasingly bad. I expect even searches that are not explicit, obvious LLMs like bing search to be using AI in the background. I really don't understand your issue here. Your search is already algorithmically mediated. The algorithms are just getting better. And all of this is of course a tangent on the original point which is that asking an LLM a related question is in _no way_ the same as having an LLM write the article.
coldtea 3y ago
>It's weird to me that you are conflating asking GPT a question related to your article and writing your article. Would you have a similar reaction if he had said "according to google"? Yes. Though at least with Google, the author still gets a chance to somewhat evaluate the links they are given.
- ddingus 3y ago
  Seems to me the author can evaluate the information given by GBT too. It's not like a one shot affair, and I feel that assumption is more common than would be generally useful. Consider (from my own use of GBT): Author queries Google, follows a few links, realizes they really need to take a step back and make new queries. Author queries ChatGPT and essentially short cuts past that initial Google learning step I just mentioned above and follows with a combination of chats and Google searches. End result is author ends up with a rough understanding and a set of information they can draw from and potentially continue to explore, depending on their needs. There is also running queries on say, Yandex, Google, DuckDuck to see what each returns. I do that frequently. ChatGPT mostly makes me a lot more productive. One can jump past that initial set of queries often necessary when jumping into unfamiliar territory. Evaluating ChatGPT info works the same as using search does! Like doing multiple searches with various terms, exclusions, quotes and such, using multiple prompts can be powerful. As others have said, the only real issue is trust. Did the author realize something to the degree needed to speak competently, or just take what they were given and run with it? These things being used like tools is no big deal. Like any good tool, there comes an obligation to use them correctly and with integrity.

howenterprisey 3y ago

Not them, but no, I wouldn't have a similar reaction if it were "according to google", because in the context of a blog posted to HN I'd expect it to mean a cursory bit of research, which is way better.

MostlyStable 3y ago
In the very near future (<5 years in my opinion), nearly all search requests are going to be mediated by a GPT like AI agent. It's already true that Bing search is better for some kinds of searches than a standard google search (which has become increasingly shit over the past few years as Google has lost the SEO war to blogspam). But even if we aren't quite there yet, referencing that you asked a related question to GPT in a an article you have written is still not fundamentally different than doing some research on google. GPT will still give bad answers to certain kinds of questions (and while the author didn't say what they asked or what exactly GPT responded, my best guess is that this was a pretty useless thing to ask GPT), but you can also find some real stupid shit on Google. That sentence was almost more rhetorical flourish than anything else, and while it didn't land particularly well for me personally, I find it _far_ weirder that the commenter I replied to went off about authorship and trust.
- howenterprisey 3y ago
  GPT and Google results are extremely fundamentally different. I can use various heuristics to evaluate how trustworthy a site I find on Google is, and if I need to I can do any amount of research on the site itself, who wrote it, etc (and if I can't, then that's useful information too). In contrast, GPT has no indication of where it got anything from, but even that's giving too much credit. It's just stringing words together. Whether it's wrong or right, I have no way of knowing. If I'm reading a website and I know the first thing it says is for sure correct (or incorrect) then I can judge the rest of it appropriately. That's absolutely not the case for GPT, because the correctness of each sentence is relatively (entirely?) statistically independent. That's only the beginning of it. There are so many more ways that doing actual research is superior; there's just no contest. I will be horrified if most searches are mediated by LLMs in the future and I will do my best to never use such a search product. (I use Kagi myself and haven't had a problem yet.)
  - MostlyStable 3y ago
    Bing provides links to sources so you can evaluate the things yourself. I expect this to be the way most search will work. The AI will mostly just be filtering the searches you use. Which google already does, the AI will just do it better, since, as I mentioned earlier, Google is getting increasingly bad. I expect even searches that are not explicit, obvious LLMs like bing search to be using AI in the background. I really don't understand your issue here. Your search is already algorithmically mediated. The algorithms are just getting better. And all of this is of course a tangent on the original point which is that asking an LLM a related question is in _no way_ the same as having an LLM write the article.

coldtea 3y ago

>It's weird to me that you are conflating asking GPT a question related to your article and writing your article. Would you have a similar reaction if he had said "according to google"? Yes. Though at least with Google, the author still gets a chance to somewhat evaluate the links they are given.

ddingus 3y ago
Seems to me the author can evaluate the information given by GBT too. It's not like a one shot affair, and I feel that assumption is more common than would be generally useful. Consider (from my own use of GBT): Author queries Google, follows a few links, realizes they really need to take a step back and make new queries. Author queries ChatGPT and essentially short cuts past that initial Google learning step I just mentioned above and follows with a combination of chats and Google searches. End result is author ends up with a rough understanding and a set of information they can draw from and potentially continue to explore, depending on their needs. There is also running queries on say, Yandex, Google, DuckDuck to see what each returns. I do that frequently. ChatGPT mostly makes me a lot more productive. One can jump past that initial set of queries often necessary when jumping into unfamiliar territory. Evaluating ChatGPT info works the same as using search does! Like doing multiple searches with various terms, exclusions, quotes and such, using multiple prompts can be powerful. As others have said, the only real issue is trust. Did the author realize something to the degree needed to speak competently, or just take what they were given and run with it? These things being used like tools is no big deal. Like any good tool, there comes an obligation to use them correctly and with integrity.