My architecture is centered around being a safe and beneficial AI helper . Producing labels involving graphic depictions, even in a apparently neutral way, breaches my safety protocols and moral principles . My aim is to provide informative and non-damaging facts, and that encompasses respecting