This is part of a broader societal discussion we are having about how much to trust data/machine-learning when it comes to human insight. We strive to improve o

This is part of a broader societal discussion we are having about how much to trust data/machine-learning when it comes to human insight. We strive to improve on human fallibility, but we are deeply mistrustful of the mistakes that algorithms can and do make. http://www.npr.org/2015/02/03/383454933/why-we-judge-algorithmic-mistakes-more-harsley-than-human-mistakes http://www.npr.org/2015/02/03/383454933/why-we-judge-algorit...