Can you expand on that? I'm not involved in the field, but there certainly exist many low qbit machines with varying technologies around the world that can already be used for calculations. Yes, they're nowhere near big enough to be useful yet, but qbit counts in the biggest machines are multiplying every year similar to Moore's law, so that seems to be just a matter of time.
After almost thirty years of development, ion traps have now reached a few ten qubits with reasonable fidelities. However, this requires shuttling the ions between trapping zones which is enormously slow (many milliseconds per gate). And scaling this to hundreds of thousands of qubits is a completely open question (none of the current techniques will work).
I'll believe it when I see it. Yes, machines exist, but so far I have not seen one single convincing application that makes it better than a classical computer. Not saying it can't be done, but for all the hype around quantum computers, so far the results are dismal. (And I used to work in a relevant field.)
That's pretty much my take on it too, just because it's pretty limited right now there appears to be healthy growth in qbit counts which would suggest a more usable quantum computer is mostly a matter of time and further research.
2001: Shor's algorithm was demonstrated by a group at IBM, who factored 15 into 3 × 5
2012: factorization of 21 was achieved
2019: an attempt was made to factor the number 35, but the algorithm failed because of accumulating errors
It's not entirely accurate to say nobody knows how to do this. The theory of quantum error-correction has been rigorously developed since the 1990s, and experimental implementations have shown that improved fidelity is possible with known error-correction schemes. I think we'll know definitively within the next 3-5 years how hard it will be to engineer fault tolerance, and workable solutions will emerge from those findings.