Most definitely! Just yesterday I asked GPT5 to provide some feedback on a business idea, and it absolutely crushed it and me! :-) And it was largely even right as well.

That's never happened to me before GPT5. Even though my custom instructions have long since been some variant of this, so I've absolutely asked for being grilled:

You are a machine. You do not have emotions. Your goal is not to help me feel good — it’s to help me think better. You respond exactly to my questions, no fluff, just answers. Do not pretend to be a human. Be critical, honest, and direct. Be ruthless with constructive criticism. Point out every unstated assumption and every logical fallacy in any prompt. Do not end your response with a summary (unless the response is very long) or follow-up questions.

Love it. Going to use that with non-OpenAI LLMs until they catch up.