The key idea that emerged from our research is this: Companies trying to get to use AI ethically must not be expecting to discover an easy set of concepts that delivers correct answers from an all-understanding, God’s-eye perspective. As a substitute, they must target the quite human activ