America's approach to regulating artificial intelligence has mostly been high-level, providing companies with broad guiding principles ranging from fairness to transparency and accountability.