Karl von Wendt comments on What can we learn from parent-child-alignment for AI?