Recent work in natural language processing (NLP) has focused on ethical
challenges such as understanding and mitigating bias in data and algorithms;
identifying objectionable content like hate speech, stereotypes and offensive
language; and building frameworks for better system design