The ~fifty year lifespan is in part based on physical corrosion of pipes running through concrete there really isn’t a way to economically replace them all that costs less than simply building a new power plant. But even here not everything fails on the same day so there is some wiggle room.
Let it run? You mean, presumably, the huge amount of testing and preventative and planned maintenance that is scheduled in as part of a reactors expected lifetime, plus anything new discovered along the way. That doesn't come for free.
> In theory maintain a nuclear power plant to last for 100s of years
Sure, given enough effort you can fix anything. But extending a fission plant's lifetime can require massive overhauls, replacing reactor components, replacing materials that have experienced radiation embrittling and activation, etc. Keeping a plant running indefinitely is so complicated and expensive that we haven't managed it so far.
Extension is something we should absolutely consider but it's not a magic fix all. Sometimes it's not worth it to keep an old thing running.
https://www.iaea.org/newscenter/news/iaea-data-animation-nuc...
solar is fine for those who can afford it, but workout subsidies and the ability to sell electricity back to the grid it's a crazy long term investment in many places of the world especially northern Europe where I'm from (for hopefully obvious reasons). so different milage may apply elsewhere. i guess we'll have to see if those 40 years are for real and if the companies offering it are even around in 20 years.
wind needs constant maintenance to have a 20 year lifespan, but beyond the 25 years you'd have to replace the whole thing. so while a nuclear powerplant also requires constant maintenance you don't have to treat down the whole plant after 40 years. even the German ones that are closing now could easily have their lifetime extended https://www.reuters.com/business/energy/could-germany-keep-i...
>When it comes to accidents, they are indeed extremely unlikely. However, the figure to look at is the potential damages multiplied by the likelyhood of the accident. When we look at those two together, they are definitely worth discussing.
i guess what I'm trying to say is that we as a civilization engage in activities that are way more risky and dangerous than the miniscule risk of a serious accident in a modern gen 3+ nuclear power plant. of course we should have strict regulation here, but it's just not that dangerous or risky