If we work hard to get more than 2 candidates from that 10%, aren't we being very biased against the 90% though? I think I'm missing some step in this logic wh

If we work hard to get more than 2 candidates from that 10%, aren't we being very biased against the 90% though? I think I'm missing some step in this logic where we should be acting in a highly biased way to get equality. If 90% of candidates are of one type, the most reasonable outcome that I can think of is if your board is 90% of that type, assuming people are largely equal. If we go off merit, and assuming there are no really significant differences between peoples gender/sex/ethnicity, it should just be a simple distribution. That's where I get really caught up with this, when people what an "even" outcome, i.e. 50% one type, 50% another (or equal percentages of each type). I think non-bias outcomes are closer to simple distribution of the candidates. Does that make sense? I'm honestly trying to bridge the gap between what I hear a lot of people's views are and what mine are.