#StopXinjiangRumors: the CCP’s decentralised disinformation campaign
3 Dec 2021| and

Video testimonials from Uyghurs saying they’re content with the economic opportunities provided for them through Chinese Communist Party re-education programs; promotion of Xinjiang as an idyllic tourism destination; commentary on the positive impact of CCP policies on the health and life expectancy of the region’s Uyghur population; content distributed in multiple languages on US and Chinese social media platforms: these are all efforts revolving around the hashtag #StopXinjiangRumors to recalibrate international perceptions of life in the Xinjiang region.

The content is distributed by social media networks that previously focused on porn and Korean soap operas but also—curiously—by CCP diplomats. Yet these networks are run by the Chinese state, directly or by outsourcing to state-directed companies linked to state and regional propaganda departments.

ASPI’s new report on Xinjiang disinformation linked to the Chinese party-state highlights how different strands of CCP online and offline information operations now interweave to create an increasingly coordinated propaganda ecosystem made up of Chinese government officials, state and regional media assets, outsourced influence-for-hire operators, social media influencers and covert information operations.

The two datasets analysed by ASPI’s disinformation team demonstrate that international criticism of CCP policy in Xinjiang continues to be acutely sensitive for the party-state and that this is driving investment in international-facing disinformation at multiple levels in the party’s propaganda apparatus.

Twitter’s state actor information operations datasets offer the most substantive evidence base for analysing the trajectory of the Chinese party-state’s online disinformation since its attribution on US social media platforms in 2019. Twitter has undertaken three major disruptions of on-platform manipulation by assets it linked with high confidence to the Chinese state through a combination of technical and behavioural signals. And Twitter is forward-leaning in making state actor information operations datasets available to the public.

ASPI’s disinformation team is one of the small handful of research teams internationally that are capable of analysing these datasets. For this report, we had advance access to the data prior to Twitter’s public release. By integrating analysis of these datasets and the CCP’s own directives and rhetoric, we learn more about how the party apparatus operationalises the CCP’s strategy for public opinion warfare.

Within the data we see overlaps that reflect different strands of pro-CCP online and offline influence activity. There are multiple intersections that suggest coordination across the party-state’s propaganda assets. Some of this is clearly directly coordinated—for example, where we see this covert information operation’s interactions with, and reciprocal amplification of, the CCP’s state and local media. Other interactions with the party’s propaganda assets, however, may be more opportunistic—for example, the engagement with prominent pro-CCP social media influencers and diplomats.

Yet cumulatively they point to the building of a propaganda ecosystem for projecting the party’s discourse power at international audiences.

The data offer insights into how propaganda directives from the top of the party structure are operationalised and suggest that there are likely to be multiple strands of CCP online information operations underway at any given time, each directed by different elements from within the party structure.

Analysis of procurement documents shows that the party is increasingly outsourcing propaganda work to a range of Chinese media, marketing and internet companies. Private-sector innovation is diversifying the CCP’s propaganda ecosystem. While publicity campaigns are generally acceptable in Western countries and not necessarily coercive, the Chinese party-state views publicity and propaganda as perfectly compatible. The datasets we analysed for our report are a good illustration of this point. They highlight how the Chinese government’s efforts to portray a positive version of Xinjiang are interwoven with disinformation denying human rights abuses.

The campaigns are also reflective of what is likely to be the future direction of the CCP’s online information operations.

Following a Politburo collective study session in May, President Xi Jinping urged the party to expand its ‘circle of friends in international public opinion’. The CCP’s recent propaganda efforts have adapted to incorporate and appropriate a more expansive circle that includes influencers and other proxies that align in projecting the party-state’s preferred narratives into international political discourse. The party is clear that this effort is an important part of its public-opinion struggle to ‘shape a more just and equitable international order and forge a new type of international relations’.

This doctrinal element adds a valuable layer to how we understand the scale, persistence and diversity of pro-CCP online influence activity. The party’s incentive structures may be driving at-scale online information operations that have performance metrics based on their ideological value on the party’s own terms—rather than on their capacity to deliver effects—as leaders of various party organs compete to demonstrate allegiance to contemporary party doctrine.

The coordination between covert information operations and other CCP propaganda assets that we identify highlights the emergence of an increasingly complex system of international-facing propaganda distribution that comprises overlapping strands of activity by diverse elements of the party-state. The CCP is leveraging asymmetric advantage in the information domain as its officials, state media and their proxies exploit the open access to international audiences that US social media platforms provide. That access, of course, isn’t reciprocal, as the CCP exercises an extensive system of control, manipulation and censorship over its domestic internet.

This report also demonstrates the value of innovative cross-sectoral partnerships and of open-source data analysis. Our industry partners at Twitter have taken the enforcement action to disrupt these assets. But ASPI identified the connection between social media accounts or channels and a local company contracted by the regional government in Xinjiang to distribute international-facing propaganda, which was an important element in Twitter’s attribution of one of these two datasets to the Chinese state. Whole-of-society responses to hybrid threats necessitate these kinds of partnerships to create the resilience we need to counter propaganda and disinformation from authoritarian regimes.