The Diplomat
Overview
The Evolution of China’s Interference in Taiwan
Associated Press, Chiang Ying-ying
Cover Story

The Evolution of China’s Interference in Taiwan

China’s foreign information manipulation and interference targeting Taiwan has been ongoing for decades. Here’s a look at its latest evolution.

By Tim Niven

On January 13, 2024, Taiwan will once again elect a president according to the country’s own constitution. Once again, Taiwan’s democracy will operate despite pressure from the People’s Republic of China (PRC).

For decades, China has engaged in foreign information manipulation and interference (FIMI) targeting Taiwan, and has optimized its tactics, techniques, and procedures, with the ultimate goal of annexing Taiwan. Beijing’s sustained, long-term FIMI campaigns affect the context of every election in Taiwan, and particularly presidential elections, where cross-strait issues dominate voter concerns.

FIMI efforts sustained over such a long timescale provide abundant opportunities for learning from trial and error. At Doublethink Lab, we have been observing and analyzing PRC FIMI targeting Taiwan for the last five years. Our observations to date suggest an evolution in tactics that appear to optimize the role of the different actors in China’s FIMI apparatus, leading to reduced risks and costs associated with attribution, while increasing effectiveness and driving societal polarization. These tactical evolutions are likely driven by the failure of China’s efforts to decrease resistance to, and increase support for, their desired annexation of Taiwan. Instead, efforts are shifting to attack the functioning of Taiwan’s democracy.

China has interfered in many democracies globally, and its tactics in those efforts also continue to evolve, causing growing international concern. In October 2023, Canada’s Rapid Response Mechanism attributed direct attacks on MPs through Facebook to “Spamouflage,” a well known network spamming over 50 social media platforms with disinformation and propaganda that has been attributed to PRC law enforcement. Microsoft Threat Intelligence identified PRC actors impersonating U.S. voters, copying Russia’s tactics, and using machine learning generated images, leading to higher levels of authentic engagement than less sophisticated previous campaigns.

China’s FIMI has been iterating not only over a long timescale, but across a broad geographic scope, multiplying opportunities for optimization. It is now a problem that all democracies are facing. Although strategic contexts differ, paying attention to developments in Taiwan, which has been called a “laboratory” for China’s FIMI, should provide productive insights for other countries.

Beijing’s interference in Taiwan’s democratic processes must first be understood as sustained psychological warfare and manipulation of the information space over decades. Psychological warfare has been persecuted through FIMI with the aims of weakening resistance to, and building support for, annexation. These efforts are supported by manipulating Taiwan’s information environment itself, including influencing traditional media, cultivation of local proxies, and collaboration with influencers.

China’s attempts to weaken resistance to annexation have, in part, focused on spreading fear, for example through military-related disinformation, and using coordinated and inauthentic behavior to amplify shows of military strength and urge surrender.

When the speaker of the United States House of Representatives, Nancy Pelosi, visited Taiwan in August 2022, Xinhua doctored a photo that purported to show a People’s Liberation Army (PLA) naval vessel visibly close to Taiwan’s coast. The fake image was republished by the Associated Press, achieving wide reach.

In their August 2023 Adversarial Threat Report, Meta identified inauthentic social media accounts on multiple platforms posting in Chinese with images and videos of PLA military might on display and a hashtag urging Tsai to surrender. Microsoft Threat Intelligence reported a similar campaign posting videos urging the Taiwanese government to surrender, in the Taiwanese language. Taiwan’s Ministry of Justice Investigations Bureau (MJIB) – similar to the United States’ FBI in remit – found that China paid a Taiwanese company to spread fear through Facebook fan pages. In this particular case, two items of disinformation were spread: that Japan would evacuate its citizens from Taiwan due to a looming invasion; and that Taiwan was “about to overstep the PRC’s red line.”

China also attempts to weaken resistance to annexation by undermining trust in the United States as a security partner. In Doublethink Lab’s 2022 election report, we identified coordinated and inauthentic behavior amplifying claims that the U.S. will bomb TSMC in the event of war, to prevent the world-leading semiconductor company from falling into Beijing’s hands.

Numerous variations on this theme – a narrative in which Taiwanese are endangered by the reckless actions of the United States – have been observed. A recent example is the claim that the U.S. will establish a laboratory in Taiwan to research biological weapons. China attempts to paint the U.S. as a self-interested hegemon using Taiwan as a pawn, insensitive to the dangers it faces, in an effort to wedge the country away from its security guarantor.

China’s tactics for building support for annexation have evolved with the times. Historically, Beijing attempted to advertise the supposed benefits of its governance model, particularly in regard to economic prosperity, which is a staple propaganda trope of PRC state media. However, we see far less of this type of content targeting Taiwan after China’s mismanagement of COVID-19, which sharply contrasted with Taiwan’s world-leading response, and has been followed by serious economic problems in China.

Taiwan remains a difficult context for Beijing’s positive propaganda, as public opinion toward China has long been unfavorable, and polling shows that the clear majority identify as Taiwanese (compared to Chinese, or both). However, a new and more insidious vector of soft power and FIMI are PRC-made and controlled apps like TikTok/Douyin and Xiaohongshu, widely referred to as China’s version of Instagram. One study has shown that, for people who are either supporters of the Taiwan People’s Party (TPP) – a self-styled "third party" led by former Taipei Mayor Ko Wen-je – or not political partisans, usage of TikTok correlates with pro-China attitudes.

We have begun to observe more suspicious activity on TikTok targeting Taiwanese, such as influencers most likely based in China spreading political propaganda and soft propaganda about the Belt and Road Initiative. Deepening understanding of China’s FIMI operations through these platforms should be a high priority in the immediate future.

China has also invested in long-term efforts to manipulate the information environment in Taiwan itself, through influencing traditional media and the cultivation of local proxies and social media influencers. Influence can be considered malign where it is either corrupt, covert, or coercive. China’s malign influence on Taiwan’s media has been extensively documented by Freedom House, and includes covert partnerships placing Beijing’s content in Taiwanese domestic news, corruption through subsidies and business ties or direct payments, and coercion through lawfare.

The cultivation of proxies both offline and online, including through United Front work, allows Beijing to pass its messages through authentic Taiwanese voices that will be far more convincing due to widespread distrust of China and its state media. Social media influencers monetizing audience attention, which have emerged as a powerful force for political influence, may also fall into the category of proxies when cultivation occurs through subsidies from China. Amplification by state media is also likely to increase the reach of pro-PRC influencers, increasing their earnings. An entire influencer economy now exists in Taiwan, which can be exploited by entrepreneurial propagandists.

The evolution in tactics that we have observed so far leading up to Taiwan’s January 2024 presidential election appears to optimize the role of the various actors in China’s FIMI system in such a way that reduces the risks and costs associated with attribution, and increases likely effectiveness. The key evolution that we observe is less initiation of information attacks, and heavier exploitation of local media, proxies, and influencers, through collaboration and amplification.

PRC state media, China-based influencers, and inauthentic accounts either internal to Beijing’s information control system or operated by proxies in third countries such as Cambodia, are far less likely to gain authentic engagement from Taiwanese audiences than authentic (or apparently authentic) local voices. By not initiating information attacks from their own FIMI apparatus, China does not open itself to the political and financial costs of direct attribution. However, by collaborating with and amplifying content produced by local proxies, Beijing is able to shape, amplify, and sustain discussion of this content, and support the pro-China propaganda economy in Taiwan. The voices amplified are also among the most divisive and act as polarizing forces in Taiwan’s society.

We also observe Taiwanese media seeding disinformation, which is then amplified by China’s state media and echoed back by pro-Beijing Taiwanese influencers. On July 9, 2023, Taiwanese newspaper United Daily News (UDN) published two articles claiming that, based on leaked government documents, the U.S. and Taiwan governments are planning to undertake research in Taiwan on biological weapons to target Chinese DNA. Almost the same pattern of disinformation was seeded by Russia to justify its invasion of Ukraine, where one variant of the conspiracy theory was that the U.S. and Ukraine were researching how to place diseases in migratory birds to target Russians.

The “Ukraine biolabs” conspiracy was embraced by China, and in the Chinese language on Twitter its spread was found to be led by PRC state media. However, the “Taiwan biolabs” fake news push was led by UDN. The leaked documents were determined by the Taiwan Taipei District Prosecutors Office to be faked, and included official-sounding phrases used in China but not Taiwan. No case has been brought against UDN, and the real source of the documents remains unknown. Tactically speaking, seeding disinformation with Taiwanese media acting as a whistleblower protected by laws sacred to democracy increases believability and decreases the risk of attribution.

We also observe PRC state media shaping, amplifying, and sustaining strategic narratives and catchphrases seeded by polarizing, high-profile influencers and political commentators in Taiwan. Shaping occurs through selective amplification, and narratives can be sustained due to the influence of China’s state media and diplomats on mainstream Taiwanese social and traditional media.

The dynamics we observe are consistent with recent work on the collaborative nature of strategic information operations, wherein elites establish “deep stories” that guide audience interpretation of events in the world (for example, “The U.S. is a self-interested hegemon using Taiwan as a pawn”), and collaborators manufacture or curate “evidence” for these deep stories (for example, the Taiwan biolabs story, or claims that “the U.S. will blow up TSMC”). The use of catchphrases reinforces interpretation and signals identity as a member of a partisan group, driving societal polarization. Entrepreneurial propagandists are able to leverage the framework provided by deep stories to create, shape, and participate in information campaigns.

A prime example of a catchphrase is the case of “hollowing out Taiwan,” referring to a conspiracy theory that the U.S. wants to relocate TSMC outside of Taiwan. In the period leading up to Taiwan’s 2022 local elections, this catchphrase was first used by influencer Julian Kuo (郭正亮) and amplified by PRC state media, subsequently becoming common. Discussion of “hollowing out Taiwan” peaked in early November, just three weeks prior to the 2022 local elections, in part driven through PRC state media efforts, starting with amplification of a professor in Taiwan who called on Taiwanese to “wake up” to being used as a pawn.

The Communist Youth League led amplification of this topic on Weibo, with the hashtag, “Taiwan professor calls on Taiwanese to wake up” (#台湾教授呼吁台湾人要觉醒#), which rose to rank seventh on Weibo’s Hot Search list, a manipulated list of trending topics on China’s version of Twitter, on November 3, 2022. The catchphrase “hollowing out Taiwan” continued to be used by PRC state media and echoed in Taiwanese political discourse in the subsequent heated discussion leading up to the election. The impact on Taiwan’s information environment was significant, with fact checkers subjected to a surge in reported rumors related to the topic.

In another illustrative case, a Russian state media journalist made a joke on Twitter (now known as X) suggesting that U.S. President Joe Biden had a “plan for the destruction of Taiwan.” Taiwan-based pro-Beijing influencer and former Kuomintang (KMT) politician Alex Tsai (蔡正元) was instrumental in this disinformation breaking out of pro-PRC echo chambers and into mainstream Taiwanese media. He quoted a “prominent radio host in D.C.” (actually the Russian state media journalist) claiming that Biden has a plan for the destruction of Taiwan in a Facebook post at 3:58 p.m. on February 21, 2023. By 6:30 p.m. the topic, “U.S. anchor reveals Biden’s plan for the destruction of Taiwan” (#美主播曝拜登曾说毁灭台湾计划#) ranked second on Weibo Hot Search.

PRC state media picked the story up on February 22, as did Taiwanese media, with some outlets laundering the narrative through PRC media. At a press conference on February 24, China’s Foreign Minister Wang Wenbin raised questions about Biden’s supposed plan. In this case, we have a disinformation campaign initiated by a pro-PRC influencer in Taiwan, laundering a narrative seeded by a Russian state media journalist, amplified and sustained by China’s state media, reaching the Taiwanese mainstream and even garnering a response from the Taiwanese government.

We also observe local proxies creating pseudo-events to give false impressions about Taiwan and generate attention to preferred talking points. A pseudo-event is one staged for the purpose of media coverage. For example, charges have been brought against Beijing’s proxies in Taiwan for taking money to organize protests against Pelosi, and paying people to cheer for a pro-China Taipei mayoral candidate. Fringe pro-unification groups organized to protest the visit of U.S. arms manufacturers also fall into this category.

Similar to the tactics above, this tactic also leverages Taiwanese citizens. Similar tactics could be used to stoke division with Taiwan’s allies, by organizing online discussions and even offline protests that can be used, for example, to paint Taiwanese society as racist against their people. Such opinions exist in all societies, but can be weaponized by such tactics. The organization of pseudo-events takes place in part offline, making such a tactic much harder to monitor and address.

The evolution of China’s FIMI targeting Taiwan may ultimately be explained by the effectiveness of different tactics, techniques, and procedures. Fear-mongering has not led to Taiwanese surrender, even though it has undoubtedly affected political discourse. Voting patterns nevertheless suggest the majority support candidates who are committed to the defense of Taiwan. China has apparently given up on failed efforts to promote its governance model. Public opinion toward Beijing, and preference for annexation, remains low. Information attacks originating from inauthentic social media accounts often fail to achieve significant organic engagement.

On the other hand, authentic Taiwanese voices – in particular journalists, local proxies, and social media influencers – are able to more readily influence Taiwan’s political discourse. The breakout of disinformation campaigns within Taiwan beyond pro-China echo chambers can be driven by influential voices in Taiwan in collaboration with state media.

By amplifying these highly divisive voices and strategic narratives, China drives polarization in Taiwanese society. Polarization may be an important tactical objective for Beijing, attacking the fabric of Taiwan’s democracy, and frustrating clear and effective policymaking prior to invasion.

Although every country’s strategic situation is different, it is possible that other countries may increasingly see a similar evolution in tactics. Heavier reliance on offline relationships that do not leave as many online traces is a difficult problem for democracies to address.

Democracies need to increase their efforts at anti-infiltration, including through legislation such as foreign agent registration, and enforcement of such laws, in order to have a chance at addressing China’s FIMI tactics. With tactics increasingly relying on collaborative efforts among a mix of authentic and inauthentic local actors, the coordinated inauthentic behavior framework will remain insufficient for platforms to address contemporary FIMI campaigns.

Policy is urgently needed to incentivize or compel platforms to increase their investment in proactive measures to combat FIMI. Public education campaigns are necessary to raise media literacy, and awareness of China’s evolving tactics in order to blunt their effectiveness. We must also figure out how to deal with the rise of influencers and the propaganda economy. Disinformation entrepreneurs such as Alex Tsai have the means and motivation to drive information attacks on their own initiative, monetizing the attention of their audiences, creating disinformation that provides evidence for “deep stories” set by PRC state media. Democracies must continue collaboration to fight FIMI, build capacity together, and gather the evidence required for effective policymaking.

Democracies should take heart from Taiwan’s example. Despite sustained and well-resourced psychological warfare, Taiwanese democracy continues to function in defiance of its authoritarian neighbor, and remains the threat of a bad example in the eyes of China – that democracy can and does work in the region. Nevertheless, democracies must remain vigilant in the face of China’s evolving tactics of manipulation in order to safeguard their democracy, liberty, and sovereignty.

Want to read more?
Subscribe for full access.

Subscribe
Already a subscriber?

The Authors

Tim Niven is Research Lead at Doublethink Lab in Taipei, Taiwan. Doublethink Lab focuses on building democratic resilience to authoritarian malign influence across multiple domains including information manipulation. Its flagship project is the China Index.

Letter
Letter From the Editors
Leads
Bangladesh’s Make-or-Break Elections
;