China's most popular messaging app has been censoring key words about the coronavirus outbreak from as early as 1 January, a report has found.
Toronto-based research group Citizen Lab found that WeChat blocked combinations of keywords and criticism of President Xi Jinping.
The report also found that WeChat, owned by Chinese firm Tencent, blocked more words as the outbreak grew.
China has for years censored what its people read and say online.
But this report suggests China began censoring discussions weeks before officials began acknowledging the severity of the outbreak.
It was on 31 December that China first alerted the World Health Organization to an outbreak of a new coronavirus in the city of Wuhan.
But authorities initially withheld information from the public - under-reporting the number of people infected, downplaying the risks, and failing to provide timely information that could have saved lives.
It was only on 20 January that Chinese president Xi Jinping publicly addressed the issue of the virus, saying it had to be "resolutely contained".
It's not clear if the social media platforms blocked these keyword combinations based on government directives - or if it was done of its own accord.
However, the report suggests that it could be the result of companies "over-censoring in order to avoid official reprimands".
Authorities have confirmed more than 92,000 cases of the virus worldwide - of which more than 80,000 are in China.
The keywords censored
A report released by the University of Toronto's Citizen Lab on Tuesday looked into two Chinese social media platforms - WeChat and live-streaming site YY.
YY was found to have added 45 keywords to its blacklist on 31 December - which made references to the virus that was then unidentified.
These key words included the terms "Unknown Wuhan pneumonia" and "SARS outbreak in Wuhan".
WeChat was found to have censored 132 keyword combinations between 1 - 31 January. As the outbreak continued, WeChat censored 384 new keywords between 1 - 15 February.
These include keywords that referenced Chinese leaders - including President Xi - as well as neutral references to government policies on handling the epidemic, and responses to the outbreak in Hong Kong, Taiwan and Macau.
Some examples of censored combinations are "Local authorities + Epidemic + Central (government) + Cover up" and "Wuhan + Obviously + Virus + Human-to-human transmission".
References to Dr Li Wenliang also accounted for 19 censored keyword combinations.
Dr Li Wenliang was among a group of doctors in Wuhan who issued the first warnings about the virus in late December.
He was later told by police to stop making "false comments". Dr Li later contracted the virus himself and died of the disease aged 33.
The report adds that it is possible that WeChat has unblocked keywords as the outbreak continues to develop. YY is known to have unblocked certain keywords.
It is not clear what keywords, if any, continue to be censored on these platforms.
The report adds that censorship in China works through a system of "self-discipline" where companies are held liable for content on their platforms.
The censorship is particularly damaging because WeChat is such a central part of many people's lives in China - it is, in effect, WhatsApp, Facebook, Apple Pay and more, rolled into one.
App users are able to book flights, hail taxis and even transfer money - all on WeChat alone. And it's not used by individuals alone - government authorities often also release official statements on the app.
"It's appalling to see the wide range of terms, even including some non-sensitive terms, [being] censored," Patrick Poon, a researcher at Amnesty International told the BBC.
"It shows how obsessed and concerned the Chinese government is [in] trying to curb any discussion... that falls outside the official narrative.
"It's totally about social control and deprives citizens of their rights to freedom of information and expression."
Censorship is pervasive in China. Sites such as Google and Wikipedia are banned - and it's not uncommon for social media companies in China to remove content that is perceived to be threatening to social stability or the ruling Communist Party.