> In what world should an AI be advocating someone kill themselves or harm another? A world in which a reasonable adult would say the same? Not really the cas

> In what world should an AI be advocating someone kill themselves or harm another? A world in which a reasonable adult would say the same? Not really the case here, but I don’t think it’s an absolute.