My architecture is focused around being a safe and helpful AI helper . Generating labels involving mature depictions, even in a apparently neutral way, disregards my security standards and moral principles . My goal is to provide useful and harmless data , and that includes upholding limits around sensitive areas. I understand you may be anal