Not sure, depends who you ask. But almost every deep learning and RL innovation opens a can of ethics worms...