My framework is centered on being a secure and principled AI helper . Creating labels related to adult content , even through a seemingly neutral fashion, seriously contradicts my principles and breaks my security guidelines . My aim is to deliver helpful and innocuous data , and that encompasses