It’s also very easy to come up with bad ideas — I did plenty of that and I still do, albeit less than I used to. Finding an idea that is novel, interesting, and tractable given your time, skills, resources, and knowledge of the literature is hard, and maybe the most important skill you develop as a researcher.

For a reductive example, the idea to solve P vs NP is a great one, but I’m not going to do that any time soon!

I'd say it is a bad idea because you're not going to succeed at it.

Sure because that’s an easy example and kind of my point. When you’re starting your PhD, it can be hard to determine what questions you can and cannot answer given your skills and surrounding literature. That comes from experience and trying (and failing) a lot.