200,000 Chinese Twitter bots silenced over Hong Kong protests


20 Aug 2019

Demonstrators in Hong Kong. Image: Vincent Yu/AP

Twitter has revealed a huge cull of accounts, purportedly from China, for attempting to spread disinformation about Hong Kong protests.

Twitter says it has suspended more than 200,000 accounts it believes were part of a Chinese government influence campaign targeting the protest movement in Hong Kong. The company also said it will ban ads from state-backed media companies, expanding a prohibition it first applied in 2017 to two Russian entities.

Both measures are part of what a senior company official portrayed in an interview as a broader effort by Twitter to curb malicious political activity on a popular platform that has been criticised for enabling election interference in the US and around the world and for accepting funds for ads that amount to propaganda by state-run media organisations.

The accounts were suspended for violating the social network’s terms of service and “because we think this is not how people can come to Twitter to get informed”, the official said. The official, who spoke on condition of anonymity because of security concerns, said the Chinese activity was reported to the FBI, which investigated Russian efforts to interfere in the 2016 US presidential election through social media.

Protest scene with protesters throwing back tear gas at police.

Protests in Hong Kong. Image: Vincent Thian/AP

Fake news organisations

Twitter traced the Hong Kong campaign to two fake Chinese and English Twitter accounts that pretended to be news organisations based in Hong Kong, where pro-democracy demonstrators have taken to the streets since early June calling for full democracy and an inquiry into what they say is police violence against protesters.

The Chinese language account, @HKpoliticalnew, and the English account, @ctcc507, pushed tweets depicting protesters as violent criminals in a campaign aimed at influencing public opinion around the world. Twitter is banned in China but it is available in Hong Kong, a semi-autonomous region.

An additional 936 core accounts Twitter believes originated from within China attempted to sow political discord in Hong Kong by undermining the protest movement’s legitimacy and political positions. About 200,000 more automated Twitter accounts amplified the messages, engaging with the core accounts in the network. Few tweeted more than once, the official said, mostly because Twitter caught many of them using its technology.

Getting aggressive

The activity was not confined to one social media platform. Facebook said it removed seven pages, three groups and five accounts, including some portraying protesters as cockroaches and terrorists, after being notified by Twitter.

Twitter has sought to more aggressively monitor its network for malicious political activity since the 2016 presidential election and to be more transparent about its investigations, publicly releasing data about state-backed influence operations since October so others can evaluate it, the official said.

“We’re not only telling the public this happened, we’re also putting the data out there so people can study it for themselves,” the official said. State-backed media organisations are still allowed to use Twitter, but are no longer allowed to pay for ads, which show up regardless of whether you have elected to follow the group’s tweet.

Twitter declined to provide a list of what it considers state-backed media organisations, but a representative said it may consider doing so in the future. In 2017, Twitter specifically announced it would ban Russia-based RT and Sputnik from advertising on its platform.

– PA Media