Can you explain this? I’ve always been of the opinion that some solutions proposed over the years have tried to do this but fail because the trusted entity is unknown a priori. I don’t see how you can invent a trust model that is basically “trust me the user and whatever choices I have made.”