Researchers use artificial intelligence to identify racist code words on social media

Researchers at the University of Rochester have developed an artificial intelligence (AI) algorithm that can identify racist code words on social media. The algorithm is aimed to address cases in which individuals use a code that involves substituting references to communities by benign words that seem out of context, in hate filled posts or Tweets (e.g. using the words Googles, Bings, and Skypes, to represent the African-American, Asian, and Jewish communities, respectively). The algorithm involves the use of AI to ‘learn’ words that tend to go along with hateful codes, thus being able to tell with an almost 80% accuracy whether a tweet like ‘gas the skypes’ is hateful.