I think this goes too far. We measure things before we fully understand them all the time. Intelligence measures doesn’t have to be a complete theory of the mind to measure real, repeatable differences in cognitive performance.
The problem is just in what humans do with that information (bad shit)