Hasn’t even OpenAI themselves turned off the feature/API which would try and tell the likelihood that a given text was AI generated, since it was too low accura

Hasn’t even OpenAI themselves turned off the feature/API which would try and tell the likelihood that a given text was AI generated, since it was too low accuracy? If even OpenAI thinks such a thing is too low accuracy, how can a third party service like TurnItIn think they can do a better job at it?