I think part of the problem here is that she's concerned that the data is biased because the people training it are biased. The data isn't racist, researchers c

I think part of the problem here is that she's concerned that the data is biased because the people training it are biased. The data isn't racist, researchers choosing data are racist. Pointing out the data is biased is just pointing at a secondary cause rather than the primary cause. The humans are the location in the chain of causality (i.e. a learning human) where permanent change is possible.