I think it’s reasonable to be more optimistic than you put it. Qubit counts and qubit fidelity have been increasing at a remarkable rate. Just five years ago we

I think it’s reasonable to be more optimistic than you put it. Qubit counts and qubit fidelity have been increasing at a remarkable rate. Just five years ago we could barely eek out a handful of qubits, and when we did, they’d be bad. Google’s quantum supremacy result is a testament to that. (At this stage, whether they actually demonstrated the “supreme” part of supremacy is, imho, irrelevant. They’ve demonstrated a much larger, controllable, programmable quantum computer and measured its quality characteristics accurately.) Of course, I’m saying it’s reasonable to be more optimistic, not that we have a proof we will certainly be able to scale to enormous machine sizes. But it’s definitely more than “speculative physics”: real machines have been built and demonstrated to exhibit truly measurable quantum effects that allow for programmable computation. (To be sure, there is hype, there is a lot of cash sloshing around, and there are totally bogus claims some companies are publicly making.)