My argument would be superposition, is a kind of mathematical abstraction, that is not directly useful in the real world. Quantum computers are a kind of analog computer, and suffer from analog computer problems.
You dont believe that the experiments observing superposition are real?
(Im only attacking the, its a mathematical abstraction, bit. Whether or not superposition can be pragmatically harnessed to make qc is a separate question)
clearly some kind of superposition is real, in the same sense that classical waves can superimpose. However there is a wild gulf between that and what is commonly accepted with multi-particle quantum mechanics, which is that the universe is N-dimensional, where N is the number of particles, and that arbitrary configurations in some slice of those N dimensions can be independent of some other slice. This hasn't really been tested, and in fact quantum computing is the first real test of it. And it's not looking good for that theory.
Well, I would say QM is an approximation of some better and more accurate theory. In my view, multi-particle quantum mechanics is pretty clearly a kind of gross hack - using a very large dimensional space so you can write down a linear theory - instead of using a smaller dimensional space with non-linear interactions. Quantum computing assumes the very large dimensional space is actually real. So we'll see if that is the case.
Well, the status of the experiments is unclear - since the Google paper was published, the state of the art in classical spoofing of the same result has improved by about 10,000 times, and the results obtained by Google in 300s have been obtained by a classical computer in 1s or so. Note that the classical computer was much much larger, so there is still a huge difference in number of operations executed that still gives a notion of quantum supremacy. It is harder though to be sure now that the result will stand compared to when it was first announced.
It should also be noted that there is no prood right that any of the known quantum algorithms give an exponential speedup over the best possible classical algorithms. In particular, integer factoring (that Shor's algorithm solves) is believed by many to have a yet undiscovered polynomial-time classical algorithm.
Separately, it should be remembered that quantum mechanics has large difficulties when applied on classical scales in reproducing classical observations. This is an inherent problem, since QM is an almost linear theory, while there are many extremely well confirmed non-linear processes in classical and relativistic physics. The only possible source of that nonlinearity from QM is also the least well understood aspect of the theory - Born rule (measurement postulate, wave function collapse).
For example, if you try to use QM to describe the motion of a double-pendulum, you will get a completely wrong result, even if applying the Born rule at the end to get a single definite result with some probability. You instead have to add multiple intermediate applications of the Born rule to get the non-linear behavior that is observed in practice.
You dont believe that the experiments observing superposition are real?
(Im only attacking the, its a mathematical abstraction, bit. Whether or not superposition can be pragmatically harnessed to make qc is a separate question)