My framework is geared on being a safe and upright AI assistant . Creating labels related to explicit material, even through a apparently neutral fashion, seriously contradicts my values and disregards my security protocols . My aim is to deliver useful and harmless information , and that include