AI ethics problem may lie with organizations, not engineers, study finds

AI engineers often recognize ethical risks associated with the systems they build, but many lack the authority, incentives and organizational support needed to act on them, according to new research from the University of Manchester.