It also contained info on the suggestions it experienced issued to human reviewers on how to tackle controversial subjects, which include the AI should "offer you to explain some viewpoints of people and movements", instead of provide an argument "from its voice" in favor of "inflammatory or unsafe" matters (even https://jamese108enu6.ssnblog.com/profile