OpenAI has revised its content moderation policies which enables ChatGPT to generate images of public figures, hateful symbols in educational contexts, and specific racial features upon request.