The Cyberspace Administration of China (国家互联网信息办公室) is a core national regulator for online information governance, and in the digital-human domain its influence shows up less as an “industry promoter” than as a rule-setter for how digitally embodied, AI-generated personas can be deployed at scale without undermining content integrity, consumer protection, and public trust: for example, it has jointly issued live-stream e-commerce supervision rules with the State Administration for Market Regulation that explicitly bring AI-generated live-streaming content such as “digital human anchors” into the compliance perimeter, including platform-side obligations around identity checks, ex ante compliance review, and management of “digital human” use (with the regime taking effect on February 1, 2026, as reflected in multiple contemporaneous notices). In parallel, CAC has co-authored horizontal “labeling/identification” requirements for AI-generated and synthetic content with agencies including the Ministry of Industry and Information Technology, the Ministry of Public Security, and the National Radio and Television Administration, and has published implementation-oriented material indicating that virtual digital humans and related interactive AI applications are expected to carry effective identifiers that reduce misrecognition risk, with technical support roles described for the National Internet Emergency Center. Finally, CAC’s regulatory “edge” on digital humans is expanding from output labeling into interaction governance, illustrated by its publication of a draft “anthropomorphic interactive AI service” interim framework (December 27, 2025) that directly targets services designed to engage users through human-like conversational and affective behaviors—an umbrella that commonly includes many digital-human products—while also signaling interest in standardized identity/label taxonomies that explicitly name “digital human identity marking” as a candidate national-standard track.
At the subnational level, Cyberspace Administration of China operates through a nationwide network of “local cyberspace departments” (地方网信部门) that correspond to each province, autonomous region, and directly administered municipality, plus a parallel office for the Xinjiang Production and Construction Corps (“兵团”) shown alongside the provincial-level links on the official CAC “cyberspace affairs” directory page. In practice these local organs are commonly organized as the standing office for the local Party cyberspace commission (网络安全和信息化委员会办公室), while also carrying an outward-facing government nameplate as an “Internet Information Office” (互联网信息办公室) with administrative enforcement functions; a concrete municipal example is Guangzhou’s commission office, which explicitly states that it “externally” uses the Guangzhou Internet Information Office nameplate and undertakes citywide internet information content management and coordinated cybersecurity/informatization supervision. Functionally, CAC-issued or CAC-hosted regulatory instruments then allocate oversight tasks down the chain: the 2016 Internet news services provisions assign supervision within each administrative area to “all levels’ Internet information offices,” with specific provincial-level roles such as organizing staff training on CAC’s behalf, and the December 27, 2025 draft rules on anthropomorphic interactive AI services (a category that often overlaps with digital-human style products) define “local cyberspace departments” as responsible for governance in their regions while requiring certain providers to submit security assessment reports to the “territorial provincial cyberspace department,” which in turn must conduct annual written review and verification.