>In 2019, algorithms from both companies were included in a federal study of over 100 facial recognition systems that found they were biased, falsely identifying African-American and Asian faces 10 times to 100 times more than Caucasian faces.
This is damning.
This story really needs a deeper investigation to how the algo works and what it uses to establish conclusions
tl;dr: Faulty facial recognition match led to a Michigan man’s arrest for a crime he did not commit.
Of all the emerging tech around us, this is way up there on what has me worried. It's potential for abuse is staggering.