I just don't think cryptocurrencies can solve real world problems. The real world is run by completely different people (people using 20th century social technologies, not 21st century ones). [...] In the real world, everything is based on laws and regulations.
I think I get what you mean but I meant something different with "real world problems". The "real world problems" I was talking about are, imo, necessities of people which are not related to the cryptocurrency sector. They can exist even in the virtual space, but they are real necessities.
"Transferring value without being censored" is such a "real world" necessity which Bitcoin solves, even if the transaction itself occurs in the "virtual world". My favourite example is an opposition movement in an authoritarian country. They need funding for their activities, and Bitcoin can provide an electronic means if the banks are denying an account.
Or take a stateless person who doesn't get a bank account anywhere and wants to save some money.
On Ethereum at the start there were some projects which looked interesting at a first glance. For example there was some kind of trust which allowed to pool people together to finance projects, and basically they were promoting it as an alternative to a bank credit and/or insurance-style services. I think it was
this one (Wetrust Platform). The market for such a platform should be enormous if it works. However, I never heard anything from this project again. My guess is that people learnt to game the DAO of the project, or that it was simply a scam, and I have had also doubt such a community can work.
ICO tokens to finance a startup, on the other hand, is a solution for a real world problem where Ethereum did work a bit. But ... you don't need a Turing complete scripting language for that. Coloured coins can do that trick on Bitcoin, LTC and friends since 2013 or so. Idem centralized stablecoins like Tether (which started on BTC). And the enormous amount of scams and rugpulls didn't really favour it as a platform for "serious" startups, above all outside the crypto sector.
Thus the problem for Ethereum is: how can it justify its Turing complete language which makes everything much more complex, but provides little solutions for "real world" problems?
Blockchain with a proof-of-work consensus algorithm is, first of all, a digital currency. People need decentralized money, which is why Bitcoin has become a mega-popular project.
A decentralized global computer (the original positioning of Ethereum) turned out to be unnecessary for most people. People are not smart enough for such a computer, they are satisfied with a laptop and a smartphone. In addition, governments of all countries are more or less tolerant of various centralized operating systems, centralized app stores, etc.
However, a decentralized operating system and a decentralized app store are not what governments and intelligence agencies want to see. This is too powerful an entity.
Perhaps (by the way), in the near future, artificial intelligence will learn to use this entity (for example, to create monetizable content and independently purchase computing power and electricity).