The Beijing AI Principles

Released on May 25, 2019, by the Beijing Academy of Artificial Intelligence, the Beijing AI Principles are a governance document of unusual precision. Their eight principles—harmony and human-friendliness, fairness and justice, inclusion and sharing, respect for privacy, safety and controllability, shared responsibility, openness and collaboration, and agile governance—appear at first reading to overlap substantially with Western AI ethics frameworks. The overlap is real but the difference is decisive: the sequence matters. Harmony (he xie, 和谐) comes first, agile governance last. This ordering reflects a philosophical priority claim that most Western frameworks do not make: that the systemic relational health of a society is the ultimate criterion against which AI development must be measured, and that individual rights and procedural responsiveness are important but secondary. Where the dominant AI ethics discourse begins with individual rights and moves toward aggregate effects, the Beijing Principles begin with the aggregate and treat individual protections as instruments of the collective good. The document