OpenAI cautions that its new Resource – like Many others now available – is not really foolproof. The tactic for detecting AI-published textual content “is imperfect and it will be Completely wrong from time to time,” claimed Jan Leike, head of OpenAI's alignment crew tasked to generate its programs safer. https://software-tool.com/