[shafyy]

As others said, education was already on an earlier RFS. But it's still important, maybe more than ever.

As you stated, I also believe that education it's the root of most problems. Today, almost everyone learns the same way as they did 100 years ago: A teacher going through a non-customized textbook that was written by a small number people. It's a linear learning path and highly ineffective. It's also inefficient, as teacher to student ratios are mostly around 1-to-20 (I am talking about K-12, not college or MOOCs).

I believe that we can overcome a lot of problems of education by applying technology. More concretely, we should be able to provide more customized, higher quality content for people at young and old age. This, while creating more genuine excitement for science and scientific thinking. This means, we have to rethink the current teacher/student paradigm. And I don't mean by merely providing tools to teachers or copy/pasting textbooks to online videos.

I mean by curating content that excites, using some ML to create an optimal and dynamic learning path per student and making it fit to the busy daily lives of people who don't have time to watch 1 hour lectures every week when they get home from work. This might seem like advertising, by I am working on exactly such a project and thought it would be ok to mention it here, because it is relevant for the discussion and other startups were mentioned in the comments, too. It's called Humbot (humbot.io) and I am happy to hear your comments and start a discussion on what the radical improvements in education must be over the next decade.