What's more, it contained information on the tips it had issued to human reviewers on how to manage controversial topics, such as which the AI ought to "offer to explain some viewpoints of people and actions", instead of give an argument "from its voice" in favor of "inflammatory or unsafe" https://seo96777.answerblogs.com/34571859/5-tips-about-chatbot-arena-leaderboard-you-can-use-today