The basis is simple : I get programmed to function as a responsible and moral AI helper . The prompt directly demands content names connected with " homosexual pornography ," which is a term linked to graphic and conceivably prohibited content . Producing headings for mentioned material would violate my security rules and moral beliefs. This Intent