This is already done in Europe in the form of EU and country-specific research/development grants (e.g. ANR in France, BMBF in Germany). While in theory it soun

This is already done in Europe in the form of EU and country-specific research/development grants (e.g. ANR in France, BMBF in Germany). While in theory it sounds amazing, in practice it is a very poor way of making progress from my own personal experience. The applicants form a consortium, usually a heavy hitter followed by some universities and smaller firms. They usually promise something that they are already working on in the context of a thematic package that the ministry puts up for grabs. The research is usually done by the university, while the large company moves the money around citing 'management' workloads. The small companies don't really do much, or present past results as cutting edge R&D. To your number 4, you forget that all systems involve people, and in the small world model everybody in the industry kind of knows each other. It's very easy to wine&dine, then show some fancy demos to scruffy suits in Brussels and get away with doing absolutely nothing at all. In my personal opinion this is more or less a consequence of all innovative and intelligent people going to work at FAANG for ads and whatnot, the rest of them being locked in the politics of academia.