https://www.thenewatlantis.com/publications/saving-science

Do we actually need a lot of science? A lot of the low hanging fruit has been picked and it seems to me like the next advances will require deep reflection and study that isn't compatible with the tenure-track-driven-by-grant-receipt / grant-receipt-driven-by-performance-on-other-grants cycle of government driven science. At the same time, if government does anything else the scientific endeavor it funds will devolve into an exercise in channeling funds to a few golden children that are selected.... How? And will be detached from any reasonable standard of good governance and accountability.

So I challenge someone to explain why the us government has a responsibility to fund science, where that authority comes from, and how acceptable standards of accountability will be established.

So the article makes the claim that left to its own devices, science wanders off into the weeds, spending a lot of money and brainpower on things which have absolutely no foreseeable application to the real world (e.g. stuff like string theory). FWIW, I do have sympathy for this viewpoint; while I think there is a place for purely curiosity driven blue-sky basic science, perhaps a smaller pie of the total R&D spending ought to be spent on such things.

From this you then conclude that government financing of science is bad (which goes against the econ 101 reason why the state must shoulder some of the R&D responsibility, so I won't bother to discuss that further here), whereas the article makes the opposite conclusion, namely that the state should take a much more hands-on goal-oriented approach to funding science (e.g. "get our boys to the moon before the commies").