How to make a racist AI without really trying | ConceptNet blog
amarashar's bookmarks 2018-09-21
Summary:
0.38801985560121732 This is analogous to what I saw when I experimented with analyzing restaurant reviews using word embeddings, and found out that all the Mexican restaurants were ending up with lower sentiment for no good reason.
Link:
http://blog.conceptnet.io/posts/2017/how-to-make-a-racist-ai-without-really-trying/From feeds:
Ethics/Gov of AI » amarashar's bookmarksBerkman Klein » djones's bookmarks