Internet censorship in China

China’s internet controls operate through overlapping state institutions, laws, and private platforms. The Cyberspace Administration of China coordinates much of online-content governance, while the Ministry of Industry and Information Technology, public-security organs, and other regulators oversee licensing, networks, and enforcement. The 2016 Cybersecurity Law established duties for network operators and strengthened requirements concerning data, security, and prohibited information; the 2017 Network Security Law implementing regulations and later measures expanded platform responsibilities.²

Chinese law requires many online services to remove prohibited material, preserve relevant records, and assist investigations. The regulatory framework covers politically sensitive material, pornography, fraud, defamation, extremism, and other categories defined through statutes, administrative rules, and agency directives. Because the boundaries can be broad and changeable, companies commonly apply extensive internal moderation in addition to direct government orders.

Real-name requirements are another central mechanism. Users of many messaging, publishing, livestreaming, and short-video services must register through identity information or a verified telephone number. These rules facilitate account suspension and investigation, although they do not eliminate pseudonyms or informal circumvention.

The technical core of censorship is commonly called the Great Firewall. It can interfere with cross-border communication through domain-name and IP blocking, DNS manipulation, connection resets, URL filtering, and inspection of network traffic. Research has documented both direct blocking and uneven, sometimes temporary interference, meaning that access can vary by service, location, protocol, and political moment.³

Filtering is supplemented by controls inside domestic platforms. Services such as WeChat, Weibo, Douyin, and search engines use automated detection, keyword lists, human review, account restrictions, and deletion tools. Platforms may also reduce the visibility of material without removing it, a practice often described as “shadow banning” or content demotion. The result is not a single blacklist but a layered system in which infrastructure providers and application companies share responsibility for anticipating regulatory demands.

Foreign websites and services that are frequently inaccessible from mainland China include Google, YouTube, Facebook, X, and many international news organizations. Access is not uniformly impossible: virtual private networks, proxy systems, encrypted tools, and other methods can sometimes bypass blocking, but such methods may be unstable, technically difficult, or subject to regulatory restrictions.

Political information is a major target of Chinese internet censorship, particularly material concerning the Chinese Communist Party’s legitimacy, national leaders, Tibet, Xinjiang, Hong Kong, Taiwan, corruption scandals, and collective protest. Historical events such as the 1989 Tiananmen Square crackdown remain especially sensitive. Censorship also expands during elections, anniversaries, public emergencies, protests, and crises when authorities view rapid information sharing as a threat to social stability.

Enforcement ranges from deletion and account closure to employment consequences, detention, and prosecution. The 2013 “seven-year” case involving online speech helped demonstrate that reposting or circulating information could attract criminal liability under broadly interpreted rules, while later prosecutions have involved alleged rumor spreading, subversion, or incitement. The chilling effect extends beyond directly punished users: journalists, researchers, businesses, and ordinary users may avoid lawful but ambiguous subjects.

Censorship does not suppress every controversial discussion. Studies of Chinese social media have found that criticism of government performance may remain visible, whereas content judged capable of encouraging collective action is more likely to be removed.⁴ This distinction helps explain why online debate can appear vigorous while organized mobilization is tightly constrained.

China’s system is also a form of market and information governance, not only a political filter. Domestic platforms have developed large moderation industries involving machine-learning classifiers, contracted reviewers, user reporting, and politically trained editorial teams. These practices influence commercial advertising, celebrity culture, livestreaming, gaming, and ordinary speech as well as news and dissent.

Filtering is often imperfect and contested. Users employ homophones, images, coded language, classical quotations, screenshots, and deliberate misspellings to evade keyword systems. Moderators respond with changing lists and contextual review, producing an ongoing contest rather than a fixed catalogue. Researchers have also shown that blocked content can be inferred from deletion patterns, making censorship itself a source of information about official priorities.⁴

China has promoted its approach internationally through concepts such as “cyber sovereignty,” which emphasize each state’s authority over domestic networks and data. Its model has influenced debates over platform liability, data localization, and national control of digital infrastructure, even though countries differ greatly in how extensively they apply such measures.⁵

Great Firewall

The common name for China’s combination of cross-border internet filtering and network-level interference.

Real-name registration

Rules requiring users of many online services to provide verified identity information, often through a telephone number.

Cyber sovereignty

A state-centered view that governments should retain primary authority over internet infrastructure, data, and online content within their borders.

Content demotion

Reducing the visibility or distribution of a post, account, or search result without necessarily deleting it.