As a researcher, I like the notion that there could be some incentive or mechanism to create new and original content, investigate new things, or conduct research in general. I'll group all of those under the umbrella of new content. I don't think things will work out as ideally as described here, but I can appreciate the sentiment and hope that it works out positively.

That said, the mental model that this article uses only recognizes that new content can fill in "holes" (interpolation). It also can expand the boundaries in new directions (extrapolation). That is a different and harder problem. If you distribute money to people "based on what most fills in the holes in the cheese", you really aren't expanding the boundaries of human knowledge as much as you are strengthening existing knowledge. They need to take boundary expansion into account here.

I also recognize that we know where the holes or boundaries are in many fields. They are "known unknowns". But this proposal does not take into account "unknown unknowns" --- things that we do not even realize that we don't know yet. It's going to be harder to incentive research into unknown unknowns when we don't even know what they are yet.