@benjamineskola writes "I don’t understand why people are so committed to pretending it’s good when it’s obviously bad." The problem is that it is bad, in many fronts, but not all, and that people don't see it is obviously bad when it is obviously bad, which is most of the time. If you know the answer yourself in advance, you can check whether the AI was good, and this can potentially save you time, or potentially make you spend more time checking something that you could have done yourself. The real problem is when we ask genAI's to give an answer to a question we genuinely don't know the answer, for lack of qualifications. However much you are qualified, there will remain many things you are not qualified to answer yourself, to yourself or to anybody else.