it's not impossible (from a theoretical, and possibly physical perspective) but it's likely that there are no problems which would motivate the necessary invest

it's not impossible (from a theoretical, and possibly physical perspective) but it's likely that there are no problems which would motivate the necessary investment to demonstrate a QC doing something useful which we couldn't also do in reasonable time on a classical supercomputer. Kind of like fusors: sure, you can do nuclear fusion in your garage but you're putting in far more energy than you are generating.