It has multiple PR problems, including the one you mentioned. :-)
Everyone who follows QC knows that simulating quantum chemistry could be commercially important, while breaking RSA is not. And as far as I can tell, no one cares anymore about doing tiny demonstrations of Shor's algorithm, to factor 21 into 3x7 or whatever. Over the past 5 years, the experimental interest has shifted to
(1) demonstrating sampling-based quantum supremacy (which has nothing to do with Shor's algorithm),
(2) demonstrating the building blocks of quantum error correction, and
(3) the prospect of doing useful simulations.
(A central reason for this is that, until you have a full fault-tolerant QC with millions of physical qubits, you almost certainly can't run Shor's algorithm in a way that will outperform a classical computer, even if you wanted to. By contrast, people are excited right now that they might be able to learn something new for physics or chemistry with just a few hundred good physical qubits.)
So anyway, we all know all of this, but popular expositions still like to concentrate on breaking public-key crypto because of its wow-factor (and, of course, the undisputed theoretical importance of Shor's algorithm, and its possible eventual security importance). In addition, there's often a time-lag problem, where the people working in the field have one set of concerns, and then the broader discussion is still stuck in the world of 1997.
Everyone who follows QC knows that simulating quantum chemistry could be commercially important, while breaking RSA is not. And as far as I can tell, no one cares anymore about doing tiny demonstrations of Shor's algorithm, to factor 21 into 3x7 or whatever. Over the past 5 years, the experimental interest has shifted to (1) demonstrating sampling-based quantum supremacy (which has nothing to do with Shor's algorithm), (2) demonstrating the building blocks of quantum error correction, and (3) the prospect of doing useful simulations.
(A central reason for this is that, until you have a full fault-tolerant QC with millions of physical qubits, you almost certainly can't run Shor's algorithm in a way that will outperform a classical computer, even if you wanted to. By contrast, people are excited right now that they might be able to learn something new for physics or chemistry with just a few hundred good physical qubits.)
So anyway, we all know all of this, but popular expositions still like to concentrate on breaking public-key crypto because of its wow-factor (and, of course, the undisputed theoretical importance of Shor's algorithm, and its possible eventual security importance). In addition, there's often a time-lag problem, where the people working in the field have one set of concerns, and then the broader discussion is still stuck in the world of 1997.