For the most part, we all trust that when we ask our computers a question, the answer it gives us will be correct, so two plus two is always four. For quantum computers though, which are still in development, the answers they give us may not be correct, so a verification method is needed. One way to check results would be to use more quantum computer resources, but researchers at the University of Vienna have a more elegant solution.
The researchers added error-traps, if you will, into the tasks for a quantum computer to run. These traps are calculations we already know the answer to, so if the computer misbehaves and provides an incorrect answer, we know there is a problem. By building more traps into the task, the more certain a user can be that the computer is operating correctly.
The researchers have already tested it in an optical quantum computer, which uses photons as qubits, but the technique can be applied to any quantum computer design.
Source: University of Vienna