Quantum supremacy is a big deal. It is a goddamn experimental evidence against Extended Church-Turing Thesis. If you never believed ECT (for example, all physicists seem to think ECT is obviously false) it may not matter to you, but it still is a serious claim.
Yes, Google probably can achieve quantum supremacy by slapping a few more qubits. But a few more qubits were, in fact, not slapped yet. So quantum supremacy is very much not achieved, and Google should not claim so.
It is serious, but as I said, it's also vaguely defined. We can continue quibbling over details for decades to come. Suppose Google does slap on more qubits so the time estimate goes up from 2.5 days to 25000 years. Is that quantum supremacy yet? Well, maybe if you used literally every classical computer in the world in parallel, and also used a very optimized algorithm, it would only take 2.5 years, which isn't that long if you're patient. So you could say that Google needs to slap on yet more qubits.
That's why arguing over whether it's really quantum supremacy is just not useful. People can always move the goalposts, so we might as well ban that word from discussion. The underlying fact, which nobody is debating, is that quantum computers are regularly doing tasks that are harder and harder for classical computers to.
And yes, the ECT is obviously false -- or at the very least, not obviously true. I'm surprised at how many CS people think it is some unassailable principle. Most CS courses devote precisely zero time to non-classical models of computation. If you've never studied a model of computation that could challenge ECT, then how can you be so sure of it? This is like being absolutely certain that I am the tallest person in the world, and refusing to ever leave my house to check.
I think the problem is that perfectly wonderful science has been put in the hands of marketing people. And those are the same marketing people that kept telling us L5 autonomy was right around the corner. Or more lately that we couldn't discount the possibility of a near-term AGI. Great science, lousy PR IMO.
The original Church Turing thesis concerns itself with computability and quantum computers do not violate it.
The Extended version goes on to say a probabilistic TM can efficiently simulate all realistic models of computation. Quantum computers very likely violate the efficiency claim and this supremacy result is strong evidence in support.
Yes, Google probably can achieve quantum supremacy by slapping a few more qubits. But a few more qubits were, in fact, not slapped yet. So quantum supremacy is very much not achieved, and Google should not claim so.