“Peter Yang believes it is significantly easier to achieve cross-functional alignment with AI agents than with human colleagues.”