LGBTQ+ bias in GPT-3. Understanding the risks and how to…

Content warning: Please note that this blog includes examples of toxic and offensive language generated by OpenAI's GPT-3. On its release in June 2020, OpenAI's GPT-3 represented a major breakthrough in natural language generation (NLG). With near human-like performance, GPT-3 is able to pen news articles, write complex code, and solve algebra problems -- without being explicitly trained to do so. It can even write your CV. While race and gender biases are well documented in NLG systems, research into LGBTQ bias is lacking in comparison and is thus the main focus of this blog post.