Case File

Case: How Platforms Governed Pandemic Help Information

Pandemic help posts connected patients and resources but faced rumor control, local-image concerns, and political risk.

During the Wuhan lockdown, help posts became a lifeline

After Wuhan was locked down in early 2020, residents faced shortages of hospital beds, medicine, transport, and information about missing relatives. They posted requests on WeChat, Weibo, and other platforms to find beds, oxygen, fever medicine, and family members. Volunteers forwarded messages and connected patients to hospitals, neighborhoods, and drivers. For a time, platforms performed part of the public-health dispatch function. Local governments still had to display effective control, so platforms began treating help posts as an image and political-risk problem. ^{[1] [2]}

A help post was also a live record. Posts described hospitals turning people away, queues at funeral facilities, shortages of neighborhood supplies, and secondary harm caused by lockdown. For a family, public detail was the quickest route to assistance. For a local official, each detail could become evidence that the city had lost control. That conflict, rather than medical usefulness, determined what remained visible.

Keywords, account penalties, and review created three filters

Amnesty International documented Chinese users using homophones, split characters, and image codes to evade censorship and warned that restricting expression damaged public-health communication. Citizen Lab's WeChat research found extensive filtering of COVID-related words and content, while users could not see the rules that triggered a restriction. A platform could block a keyword, reduce an account's reach, and then send the post to human reviewers who deleted it, suspended the account, or demanded a rewrite. ^{[1] [3]}

The filters expanded “not officially verified” into “must not circulate.” Local cadres feared that help posts would draw crowds or damage performance scores. Platforms used keyword lists, account penalties, and review teams to lower political risk. Volunteers therefore had to guess which words would survive, while hospitals and neighborhoods avoided admitting shortages in public. Verification was delayed, and an address or phone number that could save a life could disappear before reaching a patient.

Platform rules changed the route to assistance

When a request remained visible, it could produce a bed, medicine, or transport within hours. When it was suppressed, a family had to repost, join a private group, or rely on personal connections. Coded language and screenshot chains sometimes bypassed a block, but stripped away the original time, place, and contact details. Volunteers then struggled to tell whether a request was current. Visibility was determined by political-security labels rather than medical expertise.

Deletion also broke contact among patients, hospitals, volunteers, and reporters. Officials could treat the absence of a public record as proof that nothing had happened, while platforms presented censorship as rumor prevention. The CCP controlled both the route to aid and the memory of the outbreak, then shifted blame to an individual poster or an alleged rumor when the system failed.

WeChat monitoring made private requests part of routine governance

Citizen Lab found systematic monitoring and censorship of COVID-related terms and content on WeChat, without a complete public account of the rules, lists, or reasons for enforcement. The same infrastructure can track groups, identify sensitive words, and suppress unfavorable information after the health emergency ends. Temporary crisis censorship thus became a normal capacity of digital governance. ^{[3] [4] [5]}

When platforms know that deletion lowers local political risk and users know that a post may endanger an account or a relative, both sides censor themselves in advance. The government does not need to issue an order for every post. Compliance metrics and fear produce the same result. Pandemic help information is not merely moderated; it is reduced to fragments that can circulate only through private relationships.

What the public reports support

Amnesty International and Citizen Lab establish keyword censorship, coded language, WeChat monitoring, and restrictions on expression during the pandemic. They do not identify which level of government ordered each deletion or turn every unverified request into a proven fact. Individual cases require the original post, timestamp, platform action, and medical outcome. ^{[1] [3]}

The broader political result is direct. The CCP made information visibility part of stability and performance management, controlling circulation first and memory second when people most needed mutual aid. Platforms appeared to govern rumors, but they also hid medical failure and administrative responsibility for the Party-state.

Sources

1. Amnesty International report on COVID censorship and coded languageRestricted
2. Amnesty International report on attacks on expression during COVID-19Restricted
3. Citizen Lab, We Chat, They WatchLive

   View captured page
4. Citizen Lab research on WeChat censorship and surveillanceRedirected

   View captured page
5. Freedom on the Net: ChinaLive

   View captured page

×