It's pretty awful actually.... I sent a one liner prompt to Chat GPT(3.5) to generate an article about some technical topic, sent it to this detector it said 9

It's pretty awful actually.... I sent a one liner prompt to Chat GPT(3.5) to generate an article about some technical topic, sent it to this detector it said 99.98% real... there is no way to detect decent LLMs at the moment