3 ms·

This has a ChatGPT feeling.

• jamiegreen 4y ago
  it really does
• ftio 4y ago
  GP's comments until very recently have all been one-liners. The last few are all of a similar length and generic style. I'm inclined to agree. @dang, as you no doubt see to a far greater extent than regular users do, generative language model usage on HN is increasing, as is meta-commentary about their use (ie this thread). On one hand, I think GP has some cogent advice for OP. If the writing style were different, if it weren't transparently written by an LLM, it might deserve its top spot. But there's just something...wrong about accepting a conversation with a robot. There isn't anything inherently Bad about it, but it incentivizes a kind of community I don't want. Robots talking with robots. That's not a community at all. Meta-threads like this will become tiresome, but so will robot-generated comments. What's the right community etiquette here?
• leoplct 4y ago
  ahaha you guessed it right! I was interested to see the reactions: 1) If a discussion had started from chatGPT's comment, then it would have been interesting for the future of ChatGPT 2) If, as it was, it is discovered to be a robot then it makes me wonder if chatGPT is just yet another hype. In your opinion, will a chatGPT identifier be included in HN's spam filters to limit the noise?
  • ineedausername 4y ago
    Sure, but how accurate can it be... what if it flags legit human answers as chatgpt generated wrongly. Furthermore there are and will be other language models, how will it test against every ai in existence :')
  • tredre3 4y ago
    > In your opinion, will a chatGPT identifier be included in HN's spam filters to limit the noise? Hopefully it will because chatgpt isn't allowed on HN: https://news.ycombinator.com/item?id=33945628 https://news.ycombinator.com/item?id=33945628 .