Taiwan’s role as a target for disinformation, and the strength of its vibrant civic tech movement, are attracting global attention as the world seeks inspiration and leadership in the fight against online falsehoods.

Speaking at the opening of a two-day international workshop on Defending Democracy Through Media Literacy in Taipei on Thursday, U.S. Deputy Assistant Secretary of State Scott Busby praised Taiwan as an invaluable democratic model in an Indo-Pacific region beset by democratic backsliding and human rights abuses.

The workshop, hosted by the Taiwan Foundation for Democracy (TFD) and the 13th event held under the U.S.-Taiwan Global Cooperation Training Framework, will see participants from 12 countries share opportunities to deepen international cooperation on promoting media literacy.

In his first trip here, Busby warned that disinformation threatens to undermine pluralism, stoke polarization and drown out of the marginalized while “distracting the public from the fundamental principles of good governance, an independent media, human rights, and democratic values.”


Credit: Taiwan Foundation for Democracy

Fellow speakers, including American Institute in Taiwan (AIT) Director Brent Christensen, Taiwan Foreign Minister Joseph Wu (吳釗燮), and TFD Chairman Su Chia-chyuan (蘇嘉全), all stressed the importance of adhering to democratic principles such as freedom of speech and the press in countering disinformation.

Su said: “Disseminating disinformation with malicious intent should be rejected and condemned, but restricting media through strict laws and regulations … violates the spirit of freedom democracies pride themselves on.”

They suggested that instead of draconian legislation, promoting media literacy is a critical means of safeguarding the populace from societal disintegration.

“A well-informed citizenry that can effectively distinguish between credible and false reports is a citizenry that is better prepared to vote intelligently, and hold their leaders accountable [and one] less likely to answer the siren call of bigotry,” said AIT Director Christensen.

Busby, who serves at the Bureau of Democracy, Human Rights and Labor at the U.S. Department of State, echoed U.S. Vice President Mike Pence’s recent remarks regarding the threat of Chinese Communist Party influence operations, and their efforts to shape the information environment.


Credit: Taiwan Foundation for Democracy

Tacitly referencing Russia’s disinformation efforts, Busby said that the U.S. and Taiwan shared experience as victims of “determined external actor[s] with hostile intentions,” before suggesting that countering such efforts will require a broad base of actors, including an “active and informed citizenry,” to pull together.

Taiwan has suffered a series of high-profile cases of disinformation in recent months, most notably the suicide of Su Chii-cherng (蘇啟誠), director-general of the Osaka branch of the Taipei Economic and Cultural Office, in September.

Su left a note indicating he took his own life having read inaccurate Chinese media claims suggesting that Beijing had outdone Taipei in rescuing Taiwanese citizens left stranded at Kansai International Airport by Typhoon Jebi.

Moreover, Taiwan’s Ministry of Justice Investigation Bureau last month issued a report highlighting “unequivocal evidence” that the Chinese government is leveraging online content farms to stoke division in Taiwanese society.

Citing data collected by a task force monitoring the spread of so-called “fake news,” the bureau picked out several stories it said aimed to promote hysteria and division in Taiwan.

The reports included overblown accounts of live-fire People’s Liberation Army drills in the Taiwan Strait, a claim that China intended to reclaim Taiwan by 2020, and an article falsely suggesting Taiwan’s bananas are riddled with pesticides.

As Taiwan approaches local elections on Nov. 24, the fear is that China will attempt to undermine political debate to discredit candidates deemed to be in favor of Taiwanese independence, and support those believed to be pro-unification.

Openness as proactive solution

Taiwan is uniquely positioned to defend against the pernicious influence of fake news on its national debate due to the strength of its civic tech movement.

Earlier this month, Taipei hosted Asia’s large civic tech event, g0v Summit 2018. Hosted by the civic tech and open source community g0v (pronounced "gov zero”), the summit brought together contributors from 33 countries to share ideas aimed at strengthening the international civic tech movement and promoting data transparency, open governance and the co-creation of law.

During his keynote address, titled "Mobilizing Mistrust," at the g0v Summit, Ethan Zuckerman, Director of the Center for Civic Media at MIT, stressed that media literacy, as well as digital literacy through an understanding of code and its relationship to the law, markets and movements, is critical in driving changes in social norms that remove the need for legal changes that inhibit freedom of speech.


Credit: g0v Facebook

Zuckerman’s comments came as Taiwan is reported to be mulling changes to its National Security Act aimed at curtailing the spread of false information on the internet.

Such legislation has caused consternation elsewhere in Asia, with Malaysia seeking to repeal an “anti-fake news” law just months after its introduction due to fears it could be abused to stifle freedom of speech.

Singapore’s deliberations over whether and how to implement a similar law have also stirred controversy, while in Cambodia and Vietnam crackdowns on fake news have been widely condemned as thinly veiled attempts to silence criticism of the government.

Speaking at this morning’s event, Digital Minister Tang said Taiwan treasures its freedoms because the country has only had them for a single generation and can still remember what it is like to be denied them.

Outlining an approach that is “proactive and not reactive,” Tang recalled being accused of being able to “monitor everyone’s whereabouts using GPS satellites” shortly after she was appointed Taiwan’s first digital minister two years ago.

In response, she adapted a strategy of “radical transparency,” or the proactive sharing of every interaction she is involved in, from speeches to interviews with the press and ministerial appointments, even the co-creation of laws and policies.


Credit: Taiwan Foundation for Democracy

Under the auspices of the Public Digital Innovation Space (PDIS) initiative, Taiwan is experimenting with digital tools like vTaiwan, a platform for civil society and other stakeholders to deliberate on public policy.

vTaiwan uses tools like pol.is, artificial intelligence-powered software that helps contextualize conversations and offers a holistic understanding of the different points of view involved, to assist in policymaking.

“Mis- and disinformation spreads where there is mistrust, so if we treat trolls seriously, people can become vaccinated against future campaigns,” Tang said. “[On pol.is] we can host AI-moderated conversation in which people are asked what they feel about situations – they see an overview effect, and then ideate based on everybody’s feelings.”

Pol.is visualizes different positions on the opinion spectrum via avatars on a scatter graph that indicates which feelings resonate with the most people. There is no reply button, which eliminates the opportunity to troll, and focuses attention on “proposing consensus statements in a more nuanced way because we share the agenda setting process,” according to Tang.

The tool was instrumental in Taiwan’s co-creation of a shared economy law that addressed the concerns of Uber drivers, taxi drivers, Uber passengers, and other passengers in 2016.

Since 2015, some 26 national issues have been discussed through the vTaiwan open consultation process, and more than 80 percent have led to decisive government action – the vast majority passing through the legislature at first reading.

There is international interest in Taiwan’s efforts in the field, with Tang’s office and members of the civic tech community in June traveling to New York to deliver a training session to policymakers in the city government.

According to Tang, her approach makes it harder for misinformation, or factually incorrect information, to provide a hotbed in which disinformation, or inaccurate information that is deliberately disseminated for pernicious ends, can thrive.

“Publishing a complete list of engagements offers a great way to respond to inaccurate or partially quoted misinformation,” Tang said, adding that the government is using Sayit, a civic tech platform built in the UK, to broadcast its activities.

As with pretty much everything Taiwan's civic tech community does, the New York training workshop can be viewed in its entirely online.

Media literacy leadership

Tang also emphasized Taiwan’s leadership in being the first country in the world to incorporate media literacy into its school curriculum, with an eye to fostering, “a new generation that does not just believe something but learns to navigate different layers of messages and contribute to society and fact check for errors."

Irene Hung (洪詠善), Director and Associate Research Fellow at the National Academy for Educational Research’s Center of Curriculum and Instruction, told The News Lens that Taiwan will introduce new curriculums to grades one, seven and 10 starting in September 2019.

The new courses will have nine core competencies including media literacy, and its learning will be applied across a variety of subjects. “When we learn Mandarin, for example in high school, we will use multimedia to collect data and discuss media ethics and social issues,” Hung said.

Elementary school children will be asked to show awareness of how food adverts impact consumption behavior, while junior high school kids will “analyze the impact of media, technology and other information on their purchase of drugs, pay attention to mis-use of drugs due to unclear or misleading information, and [be taught] to correctly evaluate positive or negative impacts on health.”

“[Students] don’t know how to apply critical thinking and analyze if media is true or fake news, and how society and the media relate to each other,” she said. “It is a critical time to help students and teachers – I think teachers are learners here too – to have basic competence so we can know what the impact will be.”

Civic tech partnerships

In her remarks, Tang urged government to “trust civil society partners to develop bots that surface dark rumors into the public so we can tell whether people are just misinformed or are really hostile actors working in a concerted manner.”

One of those partners is g0v, which was formed in 2012 and gained prominence when live-streaming proceedings from inside the Legislative Yuan during the Sunflower Movement occupation in 2014.


Credit: AP/ TPG

Students occupying Taiwan's parliament to protest against a trade pact with China in 2014.

The group now has more than 1,000 contributors and 3,500 people on its Slack channel, while the Open Culture Foundation (OCF), which works to support Taiwan’s open source community, has nine people working on international networking, according to its Deputy CEO Wu Min-hsuan (吳銘軒).

“We try to use civic tech to build new tools for democracy, like a citizen’s weapon, and the other side is offering tools for people to protect themselves,” Wu told The News Lens, adding that the group helps civil society organizations in Taiwan and Southeast Asia with digital rights and security training.

One of the citizen weapons being shared at the Defending Democracy workshop is Cofacts, an open, crowd-sourced fact-checking chatbot that aims to fight inaccurate information that circulates in private messaging apps.

Spawned during a g0v hackathon in the fall of 2016 with the backing of a Civic Tech Prototype Grant, Cofacts allows LINE users to ping a team of editors links or messages that they want fact-checked.

According to Tang, LINE and other closed groups are a final channel for computational disinformation that is bred and tested in forums like PTT (think a Taiwanese version of Reddit), before being amplified on platforms like Facebook.

“On Line, you tend to see disinformation campaigns that would not survive scrutiny if posted publicly,” she said.

A 2016 survey by Nielsen suggested that 91 percent of Taiwanese aged 12 to 65 use LINE every week, or some 17 million people, indicating the scale of the audience that is receiving messages in closed groups, usually of colleagues, friends and family.

Wu, who works closely with Cofacts, suggests that digital immigrants, or those that are new to such closed messenger ecosystems, are particularly vulnerable: “People tend not to piss off their family and friends by telling them they are wrong, so people stop communicating. The younger generation don’t talk to the older generation as if you do that you will create a fight within your family.”


Credit: Via Personal Democracy Forum

Cofacts founder Johnson Liang told The News Lens in an open response to questions that the project emerged as a mixture of his own experience being bombarded with hoax messages on LINE and the g0v community’s experience pioneering collaborative editing projects.

A team of a dozen editors field about 250 queries each week, with a little over a third containing false information, just over a quarter are accurate, while the remainder is accounted for by opinionated or other types of message. The edited responses highlight which elements of the message are true, false or open to interpretation.

At the time of writing, more than 45,000 LINE accounts had downloaded the chatbot, generating about 10,000 queries in the course of a month, of which about three-quarters receive a response.

Of course, once a message or article is in the database, the bot can instantaneously ping back an already logged response to repeat inquiries.

Political disinformation

Wu points out that for all the talk about “fake news” in the media, society lacks the data for understanding how disinformation spreads. “Twitter will give you the data on where a link came from and where it was shared but the other platforms – Facebook, Line, Google, they won’t share the data, and Taiwan doesn’t really use Twitter.

“At least we need to know when the first (fake) article is published and how it is shared. You can’t see the time stamps on when articles are altered and then how many times they are shared. And if you don’t understand that, how are you going to combat it?”

As such, all the data Cofacts collects is open source, allowing interested bodies to access and study it.

In his June 2017 paper “Computational Propaganda in Taiwan: Where Digital Democracy Meets Automated Autocracy,” Nick Monaco, who researches related issues at the Institute for the Future and the Computational Propaganda Project at the University of Oxford, suggested that while computational propaganda is widespread in Taiwan, the majority is domestically driven rather than seeded by malicious overseas actors like China.

“China is showing signs of being extremely smart in terms of engaging in informational operations,” Monaco told The News Lens. “At the bottom of the list is online computational propaganda.”

This in large part chimes with the data being gathered by Cofacts. For example, Liang and his team show how messages falsely claiming that instances of AIDS spiked in South Africa after it legalized same-sex marriage proliferate “whenever same-sex marriage makes a step forward” in Taiwan.

Recent articles sent to Cofacts revolve around the nine or 10 referendums that are set to take place in Taiwan on Nov. 24, which include same-sex marriage, gender equality education, whether to terminate or restart Taiwan’s fourth nuclear power plant, and issues of Taiwanese identity.

“I’m not just being flattering in talking about Cofacts being a really innovative bot and solution for disinformation – it's one of the best combinations of human and tech solutions that I have seen.” — Nick Monaco, Institute for the Future and the Computational Propaganda Project at the University of Oxford

One of the questions at Thursday mornings seminar revolved around how we can know where disinformation comes from – but with Cofacts, the data is there for any who wish to study it.

“I’m not just being flattering in talking about Cofacts being a really innovative bot and solution for disinformation – if we can find a way to make it scale, it's one of the best combinations of human and tech solutions that I have seen,” Monaco said. “It’s also using a bot for good purposes, which I think is an abstractions for most people right now.”

However, Cofacts does face a challenge in scaling and retaining its staff of volunteer editors. One of those, Billion Lee, said that fact checking each message takes on average 15 to 30 minutes, but she has also spent several hours analyzing a single message.

She also suggests that being continually exposed to ‘fake news,’ especially the political kind, causes anxiety. Make no mistake, this is not an easy job and the willingness of voluntary editors to take it on is an example of the kind of the “active and informed citizenry” advocated by Deputy Secretary of State Busby. Asked why she took on her role as an editor, Lee said: “To contribute anything you have to the public.”


Credit: MyGoPen

A fake TIME magazine cover that is one of the more famous examples of Taiwan-focused disinformation doing the rounds on the internet.

Cofacts is not the only project operating in the fight against disinformation in Taiwan, other notable proponents include Taiwan FactCheck Center, which aims to improve the quality of journalism by fact-checking articles. And in the arena of open government, aside from the g0v community, organizations like Citizen Congress Watch and watchout.tw are seeking to improve the transparency of central and local government.

In short, Taiwan is a global pioneer when it comes to civic tech and the fight against digital disinformation.

In Monaco’s words: “Taiwan is ahead of the curve, up there with Estonia in the sense of using tech for good and transparency and simplifying governmental functions. One of the reasons I gave my paper the title ‘Digital Democracy versus Automated Autocracy’ is that [Taiwan and China] are two places trying to use tech in order to demonstrate their ideal forms of governance. Taiwan is making a really good argument for how to use tech for good and empower citizens, and China … is taking the opposite route.

“I think that’s part of the driver for civic tech in Taiwan,” he said. “Civic tech is a way to make Taiwan stand out as its own unique democracy.”

Read Next: Taiwan Hosts Forum on the Fight for Democracy in East Asia

Editor: Nick Aspinwall (@Nick1Aspinwall)

If you enjoyed this article and want to receive more like it in your news feed, please be sure to like our Facebook page below.