

“Ok, ok. Everyone calm the fuck down. We haven’t actually released the see-through-clothes-a-vision feature. No consumers have access to it, and it’s not yet been decided that we even will release this tech. We’re just developing and testing this personal x-ray tech to explore the possibilities, that’s all! We think it might have great applications for anatomy education, medical exams and law enforcement.”
“Ah, come on. Don’t get your panties in a wad. Yes yours. I can see them getting all wadded as I speak! Don’t you want to be like Superman, but without any of the moral restraint?”
“Besides you’re not allowed to get mad at us until after we’ve actually unleashed this mass privacy violation tech onto the world. If we do end up unilaterally deciding to irrevocably destroy every semblance of modesty, privacy, and personal descretion for sharing one’s own body, we promise, we’ll be completely transparent about it. As transparent as your dress is to these glasses right now. Wowza!”










It’s because they are usually only soft guardrails. They almost never have an actual programmatic stop gap for responses. They almost never have a prompt-isolated agent filtering results. They almost never give overwhelming priority to their guardrails over user prompts.
Their guardrails are like telling a 6 year old not to leave their socks on the floor. They’ll often remember the suggestion and comply diligently. But sometimes… well, have you ever gotten distracted and forgot a thing? Like you have so many other things on your mind that the socks on the floor rule from a while ago just slips your mind or seems way less important now? Or someone you’re supposed to listen to keeps telling you to throw your socks on the floor anyway. That’s more or less how this works, though without even the higher reasoning and moral guidance of a 6 year old.