ChatGPT/OpenAI Content Moderation System often classifies derogatory comments about Liberals as Hateful while classifying the same comments about conservatives as Not Hateful.