71–80 of 677 posts
Seems like a non problem. If it's a dull comment or just inane it'll get downvoted out of existence. If the bot actually produces something interesting about the topic, what's the problem?
The biggest issue is GPT is often confidently incorrect about things but is good enough to sound confident/authoritative so if people got into the habit of using it the signal to noise ratio would degrade.
Yes, ban it. I've been playing around with ChatGPT and where it starts failing is just where things start becoming interesting. What that means is that it's wikipedia-smart, i.e. it doesn't really tell you anything you can't find out with a minimal Google search. It does however cut the time-to-answer quite a bit, particularly if it's an area of knowledge you're not really that familiar with. But it bottoms out right…
The problem with ChatGPT is that it often reads authoritative. But is often just flat out wrong. I asked it a few questions for which I consider myself a subject matter expert and the answers were laughably wrong.
They're already banned—HN has never allowed bots or generated comments. If we have to, we'll add that explicitly to https://news.ycombinator.com/newsguidelines.html , but I'd say it already follows from the rules that are in there. We don't want canned responses from humans either! Edit: It's a bit hard to point to past explanations since the word "bots" appears in many contexts, but I did find these: https://news.yc…
It’ll be interesting if we soon come to a day when a comment can be suspected to be from a bot because it’s too coherent and smart!
#810 Constructive: https://xkcd.com/810/
The fact that soon the internet will be so flooded with bots that you'll be floating eternally alone in a sea of imposters unless we create some draconian real person ID system is a tragedy so great it's crazy it has not dawned on people yet. I started out loving the net because of the feelings of connection and partly because of the honesty and discussions stemming from at least pseudo anonymity, both silly stuff an…
The thing is you could implement a real ID system that was still pseudonymous at the user level so that you only prove real personhood at account creation. Kind of like a Twitter blue check but just for human authenticity instead of notoriety. It could even be tied back to the real person account anonymously on the admin side so you could mass ban alts automatically if people started to create bad bots. It should all…
That would be a dystopian nightmare.
Generate a Hacker News comments section discussing whether or not ChatGPT responses should be allowed on the site. Response from model User 1: I think allowing ChatGPT responses on Hacker News is a bad idea. This technology is still new, and could be used to create spam comments, or even worse, low-quality content. Plus, it could lead to more automated responses and more automation in general on the site, which I don…
Yes, ban it. I've been playing around with ChatGPT and where it starts failing is just where things start becoming interesting. What that means is that it's wikipedia-smart, i.e. it doesn't really tell you anything you can't find out with a minimal Google search. It does however cut the time-to-answer quite a bit, particularly if it's an area of knowledge you're not really that familiar with. But it bottoms out right…
They're already banned—HN has never allowed bots or generated comments. If we have to, we'll add that explicitly to https://news.ycombinator.com/newsguidelines.html , but I'd say it already follows from the rules that are in there. We don't want canned responses from humans either! Edit: It's a bit hard to point to past explanations since the word "bots" appears in many contexts, but I did find these: https://news.yc…
Dr. Casey Fiesler recently tweeted > The @huggingface GPT detector works very well on ChatGPT-created text. I ran 5 student essays and 5 ChatGPT essays for the same prompt through it, and it was correct every time with >99.9% confidence. How about adding a %human/%GPT statistic to posts and comments?
I would argue that this test isn't particularly informative. Given 5 attempts and 5 successes, even though the point estimate of accuracy is 1, the 95% CI ranges from 0.48 to 1: > binom.test(5,5,0.5) Exact binomial test data: 5 and 5 number of successes = 5, number of trials = 5, p-value = 0.0625 alternative hypothesis: true probability of success is not equal to 0.5 95 percent confidence interval: 0.4781762 1.000000…
See the app: https://huggingface.co/openai-detector/ - it gives a response as % chance it's genetic or chat bot.
They're already banned—HN has never allowed bots or generated comments. If we have to, we'll add that explicitly to https://news.ycombinator.com/newsguidelines.html , but I'd say it already follows from the rules that are in there. We don't want canned responses from humans either! Edit: It's a bit hard to point to past explanations since the word "bots" appears in many contexts, but I did find these: https://news.yc…
I disagree with this. The exact same comment written by a human is more valuable than one written by a bot.
For example imagine I relate something that actually happened to me vs a bot making up a story. Byte for byte identical stories. They could be realistic, and have several good lessons baked in. Yet one is more valuable, because it is true.