5 ms·

Bots are run against everything else, no reason to think they aren't here.

• dang 10y ago
  That theory doesn't fit the data. The overwhelming majority of flags are by users who vote and comment indistinguishably from everybody else. If they're bots, somebody should tell OpenAI.
  • paradite 10y ago
    I am very glad to know that HN has the necessary data to conduct such investigations and is doing so. Now I have much more confidence in HN. Also, it would be an interesting read if you could publish some of the findings as a post or something.
  • rhizome 10y ago
    At the end of the day, OpenAI is only providing guesses, and I don't think it's unreasonable to speculate that there could be bots that are able to simulate voting and commenting indistinguishably from other users. I'm not saying it figures in any specific thread or context, just that it's possible that bot capabilities may outstrip HN's ability to detect them.
    • dang 10y ago
      Sure, but in that case they're also passing a Turing test.