Right now the biggest problem is that it's really expensive, and takes a long time to build. Investing billions that won't start pay interest for at least 10 years isn't attractive to banks that need to show a profit every year, or governments that have to face an election every 4 years.

That's one of the reasons why small modular reactors are interesting. You don't have to invest so much before you start to see a return.

It's also why wind is doing so well. It's easy to borrow the cost of a turbine, or even a wind farm, and it starts making money pretty quickly.

The biggest scaling problems with nuclear are that we are building basically none of it right now so scaling it up means pretty much building the industrial base up from scratch, and the lack of qualified people which will need decades to train enough. It takes years to train up qualified nuclear technicians, and you'd need to scale up the schools too.

Meanwhile, the industrial base for wind and solar are massive now and increasing almost exponentially, and educating new workers takes months at most because there's nothing really complicated about them.

Recycle is 95% un-fissioned. Need either recycling infrastructure (like France but newer tech available now) or fast-spectrum reactors (Russia in the lead, god damnit).

Geological repository for what is left, or just put it all underground.

Oklo natural reactors 2 billion years ago show random geology capable of trapping “used” nuclear fuel, so picking good geology not a challenge.