Nice website and a good exercise! We really are living in a golden age when we can identify most LLM written stuff. Longer from posts are currently easier to identify than shorter. (I ended early got 4/5 The long dashes and general questions to the audience not the commenter were a tell)
Very soon we won't be able to tell any difference or identify any signs.
Some (but not all) of the AI comments are prefixed by an extra blank line. I got S (5/5), A (4/5), and A (4/5) using this tell. Took me less than a minute on the first one (as it took a few seconds to notice the blank line), the second one took 18s, and the third one took 22s.
psychosis.hn is a daily game. Every day we fetch three stories from a previous front page of HN, each with 5-7 AI comments threaded into the discussion. They have personas, reply to real people, and sometimes have real comments reparented underneath them.
First try I got 0/5 with 3 false positives, by the third i got 4/5 with 1 false positive.
it's getting scary, i don't know how much longer we can tell humans apart from AI
Nice website and a good exercise! We really are living in a golden age when we can identify most LLM written stuff. Longer from posts are currently easier to identify than shorter. (I ended early got 4/5 The long dashes and general questions to the audience not the commenter were a tell)
Very soon we won't be able to tell any difference or identify any signs.
Some (but not all) of the AI comments are prefixed by an extra blank line. I got S (5/5), A (4/5), and A (4/5) using this tell. Took me less than a minute on the first one (as it took a few seconds to notice the blank line), the second one took 18s, and the third one took 22s.
psychosis.hn is a daily game. Every day we fetch three stories from a previous front page of HN, each with 5-7 AI comments threaded into the discussion. They have personas, reply to real people, and sometimes have real comments reparented underneath them.
I've got to say, that's pretty damn good, and pretty damn scary.
Seconded. I scored 0, missing all the bots and falsely marking some real human comments.
Wish this post got more of a response, so thanks so much for giving it a try! Hope it was at least fun (and maybe a bit horrifying) :)
Responses (well, all engagement) are a lottery*, so don't let it get you down. :)
* See e.g. mine: https://news.ycombinator.com/submitted?id=ben_w