Nuclear is one of the few technologies that has a negative learning curve. As we improve designs, it seems to get more expensive rather than less.
There was a brief window in the 1970s where US nuclear construction projects were finishing on time. But the utility industry had planned for waaaaay too much new capacity. So when all the construction projects with poor execution, that struggled to complete and therefore came in way over budget, finally came online in the early 80s, they were financial disasters in a scale that nearly bankrupted several utilities.
Since then, utilities lost their appetite. And there's basically no way for us to replace the 400 or so reactors in the world that eMate nearing end of life.
However, I'm not sure we will need nuclear. With how cheap wind and solar are getting, far faster than anybody anticipated, we have finally found the technologies that may some day provide energy "too cheap to meter." However, like nuclear they are not dispatchable (except for some designs in France), so if we want to power a grid we either need to overbuild capacity by quite a bit, or use energy storage. There's a cost trade off for the two that depends on how cheap storage gets, and how cheap extra capacity is, and how cheap transmission is from an area with different weather that day. (For example, one can imagine building 2x of panel capacity over the amount of inverter capacity on a solar install, so that even on cloudy days you can chug along at near full energy output... it all depends on the cost trade offs.)
And as fast an solar is getting cheap, far beyond expectations, so is lithium ion storage. And there are many chemistries with high specific energies (and thus unsuitable for vehicles), that we are just now dipping our toes into.
Nuclear would be a nice tool to have, if it was competitive with other technologies, but it's going to be decades before it can prove itself and establish a positive track record for deployment. Utilities have been burned too many times by financial dumpster fires.
