Serious question: if I keep all known protected class information out of my pricing model, and it still decides to charge macOS users more for the product than

Serious question: if I keep all known protected class information out of my pricing model, and it still decides to charge macOS users more for the product than Windows users, is it discrimination? Is it unfair? Now if I go and look at real outcomes and add back in the protected class information that was withheld from the model and run an analysis, and the results turn out to be post-facto biased, is it discrimination? Is it unfair? Doe the FTC have an objective and consistent framework for evaluating this stuff?