setting aside alignment to humanity, does anyone have approaches for personal ai-human alignment for an agent to take custody of a human's assets & memes & genes and propagate them using their best judgment?