OpenAI has a 99.9% accurate ChatGPT AI text detector, but won't release it.ModerateImprovement@sh.itjust.works to Technology@lemmy.world – 331 points – 2 months agothe-decoder.com67Post a CommentPreviewYou are viewing a single commentView all commentsIf you believe this, I have a bridge in Brooklyn to sell youA routine that just returns "yes" will also detect all AI. It would just have an abnormally high false positive rate.My model has 100% recall and 50% precision, not bad eh? But - that model would not have 99.9% accuracy.Agreed. Personally I think this whole thing is bs.Ofc they just look in their database if this is something it has ever said and to who1 more...
If you believe this, I have a bridge in Brooklyn to sell youA routine that just returns "yes" will also detect all AI. It would just have an abnormally high false positive rate.My model has 100% recall and 50% precision, not bad eh? But - that model would not have 99.9% accuracy.Agreed. Personally I think this whole thing is bs.Ofc they just look in their database if this is something it has ever said and to who1 more...
A routine that just returns "yes" will also detect all AI. It would just have an abnormally high false positive rate.My model has 100% recall and 50% precision, not bad eh? But - that model would not have 99.9% accuracy.Agreed. Personally I think this whole thing is bs.
My model has 100% recall and 50% precision, not bad eh? But - that model would not have 99.9% accuracy.Agreed. Personally I think this whole thing is bs.
If you believe this, I have a bridge in Brooklyn to sell you
A routine that just returns "yes" will also detect all AI. It would just have an abnormally high false positive rate.
My model has 100% recall and 50% precision, not bad eh?
But - that model would not have 99.9% accuracy.
Agreed. Personally I think this whole thing is bs.
Ofc they just look in their database if this is something it has ever said and to who