I'm continually surprised that these articles on quantum computing never mention D-Wave, who have a working 2000 qubit machine. Sure, it's not a gate model, but it's working and available today, exhibits macroscale quantum effects, and is well on its way to demonstrating quantum advantage on optimization problems.

It's always the same old, 5-qubit useless IBM machine, when production, commercially available quantum machines are actually being built right now. They have a real PR problem when articles like this never mention D-Wave.