More experiments are always worthwhile, but from a practical perspective funding is a zero-sum game with a small number of finite pots. The question being raised isn't whether this proposal is worthwhile "in general", but instead whether it's a good use of finite finding vs other potential science.
Given the immense cost and the lack of serious reasons to expect otherwise, I think it's a pretty reasonable question to ask.
I've heard it expressed numerous times that the zero-sum-game argument is a fallacy. There are a lot of foundations and donors that pick among projects and might just hold on to their money if a suitable project to fund is not available.
I can not comment on the amount of money, but you do get that the question to be tested by this proposal is the main unsolved question that bothered Schroedinger, Bohr, Einstein, and others, right?
The $1B pricetag came from the linked article. The thing about the main unsolved question is simply false - the process by which classical mechanics emerges from quantum mechanics is already well-understood and partly taught in undergraduate courses.
Yes, I have even been an instructor for some of these classes. No, knowing how to take a limit / apply steepest descent does not tell you whether your model of reality is correct, it just lets you convert your model (quantum) to a more restricted model (classical). But you do not know if your intial model is not itself the limit of another model.
Converting the quantum model into a classical model as the limit is taken of greater distances is an explanation of why we don't see quantum mechanics in our daily lives. It shows that no extra dynamics are necessary to "kill" quantum effects on macroscopic scales, because within the laws of quantum mechanics, it is already explained.
Decoherence theory explains that if you have an entangled system that is not sufficiently isolated it will pull in the environment (including the researchers) quickly into the superposition. This explains why a human won't see a "blur" or whatever if you look at Schrödingers Cat. But this seems apparent from the start unless you subscribe to the Copenhagen interpretation literally (something I'm pretty sure not even most of the earliest QM researchers did).
Since decoherence is just a linear (normal) process in the linear QM realm, you still need to postulate the Born rule and subscribe to one of the interpretations thereof, even if you're an Everettian, to make predictions. I think this is the main open ended question in QM..
But sure, there are many fringe theories going around and doing tests to falsify them would be good, it just seems steep to do it at $1B and the experiment only pushes the limit quantitatively, not qualitatively (as the same experiments are done on Earth regularly with lighter objects).
You're just moving forward the limit a bit by the proposed experiment - if it works (like is generally assumed I think), the sceptics will just say that the breakdown limit is higher than the mass of the glass-beads.. :)
Given the immense cost and the lack of serious reasons to expect otherwise, I think it's a pretty reasonable question to ask.