too much to chalk it up to bureaucracy.
I'd blame it on a lack of experience and a lack of economy of scale.
If you want to see nuclear power get lower in cost then create a means to have economy of scale and plenty of experience. France tried this by building a bunch of the same reactor all over the country. That worked until they found a flaw common in all the reactors that cost a lot to fix, and they stopped building them meaning they lost all their experienced workers.
Let's try this, and a nation like the USA is one of the few big enough to try. Start with as many nuclear reactor designs that look like they might work. I don't mean go crazy and pick 100, or pick only 2. Perhaps 12. Get companies to build one of each. After they finish evaluate them as best you can on costs, safety, and so on. This is going to be quick-n-dirty because until they run for a few years there's still plenty of unknowns. The worst 1 or 2 of them aren't built any more. This makes the one and done reactors expensive to run but it's finished, (presumed) working, and so will still pay itself off in time. The remaining are then repeated, but this time build 2 of each. Evaluate again, drop a design, build 4 each of those remaining. After a while there will be something like 3 or 4 very solid designs of nuclear reactors on the market. Each had to compete on costs and safety, and there's experienced crews to build more at low costs. Costs will stay low because the competition remains, people will simply gravitate towards the best value.
Because the USA has a market for potentially as many as 1000 gigawatt scale nuclear reactors, and new reactors are likely to be smaller than the old gigawatt ones, this means a lot of people can keep building nuclear reactors to maintain experience. And the market might settle on just 2 or 3 designs for a while but there needs to be new designs introduced once in a while to allow for innovation. It's new designs that incorporate what was learned in building the old designs that will lower costs more than anything. But it will take years of building what are just moderate updates from designs out of the 1990s to get the experience we need to innovate.
Last i heard nobody did jail time for Fukushima .
Why should they? The reactors exceeded their government required specifications for earthquake survivability. The wall to contain the tsunami met government spec. This was a tsunami that was not expected to ever be seen. The people on site acted admirably. There was only one suspected death from radiation, the victim was well cared for while ill, and the family compensated for the death. The people working there knew of the dangers. Other deaths on the site were from the flood, not the reactor. Again, any fault on flood victims will be difficult to pin on anyone because they met government safety requirements.
Deaths from the evacuation can't be blamed on them since they can argue the evacuations were unnecessary given the levels of radiation beyond the borders of the power plant.
There's people that die from industrial accidents, storm damage, and so on at solar power facilities, wind farms, hydroelectric dams, or whatever else you can think of as an energy source besides nuclear. I don't see these people going to jail. Not every death is the result of a crime. That's not blaming the victim either. Accidents happen where there's no way to find fault with anyone.
I keep having people bring up that nuclear power just costs too much to try. Well, if that's where it ends then we'd never get anything new. Go back to the drawing board and try again. Don't make it a half-assed effort in building anything new. Give it an honest chance to prove that the design can come in on time and budget. It can take three or four budget busting efforts to work out all the problems. If global warming is half the problem that it is claimed to be then it is in our best interests to give nuclear power a try.
Oh, and there is no one nuclear reactor. If one reactor project goes over budget that should not reflect on different project. This is no different than claiming a 2020 Tesla S is unsafe because of what was seen with a 1972 Ford Pinto.