> It is not obvious that nuclear energy can be a widespread, long term alternative for most of the world economies.
This problem is also known as nuclear proliferation and, as far as I know, is known to people building nuclear reactors. Small nuclear reactors takes them in account: https://en.wikipedia.org/wiki/Small_modular_reactor#Nuclear_...> ; nuclear smuggling is also monitored: https://www.iaea.org/newscenter/pressreleases/iaea-database-...>
Unpopular opinion: I do not think it is possible to get rid of the inherent risk of nuclear proliferation, even without building nuclear reactors in non-wealthy countries. Multiple reasons :
* Non-wealthy countries wants to be first-class countries, at some point, they will desire to move on to "energy upgrades", beyond international community (read: wealthy countries), who has the right to refuse them to build such infrastructure?
* Villains will move on to harder targets, but will move on, even if it requires using countries already using nuclear and are actually "non-wealthy" (for some reasonable definition)
It might not be feasible to guarantee the right levels of security for these scenarios, but we can take them in account and work out nuclear reactor models which limits as much as possible the inherent risks.
Consider this bad metaphor: as we are writing more and more software, some of them are increasingly critical to users, therefore, we invest in different levels of guaranteeing software is working as intended (e.g. formal methods, etc.).
I see no absolute reason to not pursue the same avenue here, indeed, I even learned about nuclear reactor startups which are applying modern technology to a safer nuclear ecosystem.