>Well... I live in a world in which 99% percent of fire fighters are male

So.. Not this world, because it's more like 20% here. (And would be bigger if females weren't harassed so much.)

>no-one made a big deal out of a sex disparency in fire fighting

Sure, ignore history. You are doomed to repeat it.

> I am designed an AI algorithm to select who will be best at 'task X' I wouldn't call the algorithm biased/poorly coded if it overwhelming selected from the group shown to be better suited for task X.

Good thing nobody asks you, because that is the wrong algorithm. Maybe it's plausible short-cut if you are looking for "the best in the world". But given an arbitrary subset of people, it's not always going to be a male winner. You suck at writing algorithims.

>I haven't 'addressed' it because I think the statement is markedly untrue.

Let's summarize so far, shall we?

- You asked how an AI could be racist. I gave you links. You ignored them.

- You asserted the AI is not biased (without any evidence), and later doubled-down by saying those articles are "untrue" (again.. without any evidence)

- You claimed that 99% of firefighters are male (without evidence)

- You assert that "picking all males for a SW position is fine" (without any evidence, and despite me pointing out that it is literally illegal), then doubled down implying that you personally would preferentially hire only males even though there is no evidence that males have an advantage in SW.

You are blocked.