Hrm, I think I understand what you're saying. Perhaps a better way to formulate it is: let's say we take a model and add a regularization term that penalizes/r

Hrm, I think I understand what you're saying. Perhaps a better way to formulate it is: let's say we take a model and add a regularization term that penalizes/rewards some part of the data, which when trained makes for a morally/ethically biased model. I think you're exactly correct, the source of moral/ethical bias here is in the model. Good point.