Different fields, different appropriateness? Unpacking emerging normativity in China’s AI governance

How does a state initially form its stance on artificial intelligence (AI) governance before projecting it onto the international stage? Drawing on field theory, this article argues that such formation is both a pluralistic process – where diverse normativities, or understandings of appropriateness, emerge across intersecting yet distinct fields – and a dynamic one, shaped by competition among various actors seeking to influence the state’s overall approach. Using China as a least-likely case study, this article finds that different normativities of AI governance emerge across three fields: security, technology and diplomacy. These variations are driven by two factors: (1) the specific issues at stake within each field and (2) the dominant actors whose practices produce these normativities. It introduces two mechanisms to explain how these diverse normativities combine to shape China’s overarching, state-level normativity regarding AI governance. This analysis extends a Bourdieu-informed perspective on norm research, showing the complexity and fluidity of emerging normativities as well as the hierarchies and power relations that shape them. It also offers empirical insights for working with China on prospective global AI governance frameworks.