Expanding Freedom and Democracy
Freedom on the Net 2021

Taiwan

Free
80
100
A Obstacles to Access 24 25
B Limits on Content 31 35
C Violations of User Rights 25 40
Scores are based on a scale of 0 (least free) to 100 (most free). See the research methodology and report acknowledgements.

header1 Overview

Taiwan hosts one of the freest online environments in the Asia region. The information landscape is characterized by affordable internet access, diverse content, and a lack of website blocks and internet shutdowns. An independent judiciary protects free expression. Civil society, the tech sector, and the government have taken innovative action to counteract the impact of disinformation campaigns originating from China. However, criminal prosecutions for online activities and concerns over disproportionate surveillance all threaten internet freedom.

Taiwan’s vibrant and competitive democratic system allows peaceful transfers of power between rival parties, and protections for civil liberties are generally robust. Ongoing concerns include foreign migrant workers’ vulnerability to exploitation, and the Chinese government’s efforts to influence policymaking, the media, and democratic infrastructure.

header2 Key Developments, June 1, 2020 - May 31, 2021

  • The country boasts very high rates of internet access with no significant digital divides. Recent improvements in access are evident among previously less connected segments of the population, such as immigrants (see A1 and A2).
  • The Telecommunication Management Act (TMA) came into effect in July 2020, replacing the Telecommunications Act. The TMA relaxes previous requirements for service providers to register and retain a specific amount of capital to operate in the country, although it does impose limits on direct foreign ownership of telecom services (see A4).
  • Under a new rule issued in September 2020, Taiwanese companies could no longer provide Chinese over-the-top (OTT) services, particularly iQIYI or Tencent, via television or other broadcast, although the services remain unblocked by internet service providers. The rule updated the Act Governing Relations between the People of the Taiwan Area and the Mainland Area (see B1 and B6).
  • The government introduced the Internet Audiovisual Service Management Act (IASMA) in July 2020 to regulate OTT services. The draft law would mandate that they abide by certain transparency and registration requirements, and seeks to ensure that content on their platforms do not endanger national security, public order, or the moral good, or impair the emotional or physical well-being of youth (see B3 and B6).
  • The draft Digital Communication and Broadcasting Act (DCBA) would impose varying degrees of obligations on content providers. For instance, the draft mandates that companies disclose certain practices in their terms of service, and requires intermediaries to implement a notice-and-takedown mechanism to remove illegal content immediately after being notified of its presence (see B3).
  • At least four people were fined after being found guilty of violating the Social Order Maintenance Act for social media posts that included inaccurate information (see C3).
  • The Technology Investigation Act, proposed in September 2020, would increase authorities’ ability to monitor private communications stored on a user’s electronic device (see C4 and C5).
  • In May 2021, the government introduced the 1922 SMS contact-tracing system, which uses QR codes to track users’ locations (see C6).

A Obstacles to Access

A1 1.00-6.00 pts0-6 pts
Do infrastructural limitations restrict access to the internet or the speed and quality of internet connections? 6.006 6.006

In general, there are no infrastructural limitations to internet access in Taiwan and the country boasts high rates of internet access. In 2021, the social media management platform HootSuite placed Taiwan’s internet penetration rate at 90 percent, or approximately 21.45 million people.1 Other data sources placed the percentage slightly lower: the Taiwan Network Information Center (TWNIC) reported in 2020 that the penetration rate stood at 83 percent.2

Users can get online through a range of different connections, including fixed-line broadband networks like fiber, digital subscriber line (DSL), fourth- and fifth-generation (4G and 5G) technology for mobile networks, and free public Wi-Fi services.3 According to the National Communications Commission (NCC), 5.83 million people in 2019 subscribed to fixed-line broadband networks,4 while the penetration rate for mobile networks was 114 percent.5 There are nearly 10,000 hotspots across the country providing free Wi-Fi.6

The government is dedicated to upgrading mobile services to 4G networks and promoting 5G.7 2G was suspended in 2017,8 and telecommunications companies stopped offering 3G contracts the following year.9 Chunghwa, Taiwan’s largest telecommunications company, has announced that it will stop providing 3G technical assistance for calls by 2024.10 The major telecommunications companies, such as ChungHwa Telecom, Taiwan Mobile, and Far EasTone (FET) have begun providing 5G service in major cities and several other locations throughout the country.11

In July 2021, Taiwan was ranked 28 and 30 in Ookla’s SpeedTest global index for mobile and fixed-line broadband speeds, respectively. That month, Ookla reported Taiwan’s average mobile data download and upload speeds as 71.82 megabits per second (Mbps) and 15.09 Mbps, respectively, with fixed broadband download and upload speeds at 149.83 Mbps and 88.94 Mbps.12

Taiwan Academic Network (TANet), which is maintained by the Ministry of Education and several universities, provides the network infrastructure for educational institutions, including universities and libraries.13

A2 1.00-3.00 pts0-3 pts
Is access to the internet prohibitively expensive or beyond the reach of certain segments of the population for geographical, social, or other reasons? 3.003 3.003

There are no significant digital divides in Taiwan, although slight disparities remain based on geographical area and age. Internet access, especially mobile networks, is affordable. According to the report from the TWNIC, nearly 95 to 97 percent of users spend less than 1 percent of their monthly income for mobile networks.1 The Inclusive Internet Index 2021 report ranks Taiwan 47 out of 100 countries in terms of affordability, defined by cost of access relative to income and the level of competition in the internet marketplace.2

There is a slight digital divide with regards to geographic location, although it has been diminishing in recent years.3 A 2020 report from TWNIC stated that 84 percent of people above the age of 12 in nonrural areas had access to the internet, compared to only 70 percent in rural areas.4

There is also a slight disparity in access based on age. In 2019, the National Development Council (NDC) reported that 86.2 percent of people above the age of 12 accessed the internet, but only 75 percent of people between the age of 60 to 64. Only 44 percent of people over the age of 65 used the internet.5

There is not a significant gender divide to access. Some surveys report that men use the internet more than woman by a rate of about 2 to 5 percent.6

Other groups have experienced a boost in internet access in recent years. For example, as of 2020, 96 percent of immigrants used the internet, a sharp increase from 72 percent in 2014.7 The government has also set up the i-Tribe program to increase wireless broadband for Indigenous communities.8 The program has reportedly improved people’s ability to access digital health care services and other information.9

A3 1.00-6.00 pts0-6 pts
Does the government exercise technical or legal control over internet infrastructure for the purposes of restricting connectivity? 6.006 6.006

The government does not intentionally restrict connectivity, and the country’s internet infrastructure is privately owned.

Taiwan’s four internet exchanges points—TWIX, TPIX, EBIX, and TWNAP—are all operated by telecommunications companies, although TWNAP works largely only as a data center and not an exchange point.1 The submarine cables connecting international networks are also privately owned.2 ChungHwa Telecom, 35 percent of which is held by the Ministry of Transportation and Communications (MOTC), lays the majority of submarine cables.3 In 2020, Google and Facebook also proposed a plan for a submarine cable to connect the United States and Taiwan.4

A4 1.00-6.00 pts0-6 pts
Are there legal, regulatory, or economic obstacles that restrict the diversity of service providers? 5.005 6.006

While users have a choice of providers for internet services, certain companies dominate the market. The new Telecommunication Management Act (TMA),1 which was approved in June 2019 and came into effect in July 2020, replaces the previous Telecommunications Act (TA), and relaxes some of its rules. There is a three-year transition period for telecommunication companies to comply with the TMA.2

Previously, under the TA, companies that “install telecommunications equipment or provide telecommunications services” required a license from the Ministry of Communications and to obtain a specific amount of capital.3 However, the TMA loosened these requirements, meaning service providers must now only register.4

Under the TMA’s provisions, direct foreign ownership of telecommunications services is limited to no more than 49 percent, and only 60 percent of shares may be owned indirectly or directly by foreigners.5

The TMA places some obligations on service providers, though they are not particularly onerous, and are often meant to protect consumers. For example, telecom operators must take appropriate measures to protect the confidentiality of communication, provide public and easily accessible information to consumers, separate telecommunications and service fees from unrelated ones, and provide channels for consumers to lodge complaints.6

Previous requirements to enter the market and the high cost of developing infrastructure, among other factors, allowed only a small number of providers to dominate the fixed and mobile market.7 Five major telecommunication companies—Chunghwa Telecoms, Taiwan Mobile, Far EasTone Telecommunications (FET) Asian Pacific Telecom, and Taiwan Star Telecom (T Star)—occupy the majority of the market,8 with Chunghwa Telecom controlling approximately 68 percent.9 ChungHwa Telecom, Taiwan Mobile, FET, T Star, and Asian Pacific Telecom also provide mobile broadband service, with ChungHwa Telecom reporting the highest number of subscribers at 11.3 million as of December 2020.10 While there were 82 telecommunication companies that offered fixed-line networking as of February 2020, most were small businesses that only provide local services.

The TA classified providers that do not invest in physical lines or equipment as a “Type II” telecom operator. This allowed them to offer services through peering with other telecommunication companies, and to provide added-value amenities.11 Under the TMA, these companies only need to register instead of applying for a license.12

A5 1.00-4.00 pts0-4 pts
Do national regulatory bodies that oversee service providers and digital technology fail to operate in a free, fair, and independent manner? 4.004 4.004

Regulatory bodies that oversee telecommunications and other internet-related issues in Taiwan are generally seen as free, fair, and independent.

Established in 2006, the National Communications Commission (NCC) is an independent government body responsible for regulating telecommunications and broadcasting services, including overseeing the telecommunications industry, managing domain names and IP addresses, and processing and overseeing licenses;1 it has additionally governed TWNIC since 2017. The NCC’s mission includes promoting sound policy, safeguarding users’ rights, protecting consumer interests, and ensuring fair and effective competition in the market.2 The body is composed of seven commissioners who serve four-year terms, all of whom are nominated by the prime minister and approved by the Legislative Yuan. The prime minister is tasked with appointing both the chairperson and vice chairperson, however, prompting questions about the body’s independence.3 According to a report released by the Taiwan Public Opinion Foundation in November 2020, 68 percent of respondents over the age of 20 reported being concerned about the NCC’s independence.4

At the end of 2020, the NCC rejected an application from the pro-Beijing television channel Chung Tien News (CTi TV) for its television license renewal after the NCC had repeatedly fined and issued warnings to the channel for breaking regulatory rules. Due to the license rejection, CTi TV lost its ability to broadcast on television in Taiwan, although its operations on the internet were not affected.5 Reporters Without Borders (RSF) issued a statement saying that the de facto shutdown of a news channel was an “extreme measure,” but noted the channel’s repeated violations and stated that the NCC’s move did not constitute a violation of press freedom.6

Several other government bodies oversee digital technology. For example, the Fair Trade Commission (FTC) oversees competition law as it relates to telecommunication services. Decisions by both the FTC and NCC can be appealed to the judiciary.7 In recent years, several different government bodies have supervised the implementation of Taiwan’s Personal Data Protection Act (PDPA) (see C6). The Department of Cyber Security oversees issues related to security of critical infrastructure (see C8).

The nature of the online information dictates which agency is tasked with particular content regulation (see B2 and B3).8 For example, online content related to food hygiene is handled by the Ministry of Health and Welfare. The Institute of Watch Internet Network (iWIN), a semiofficial organization funded by several government departments, is responsible for content related to children and youth.

The Ministry of Digital Development (MDD) is expected to be operational in 2022 and is intended to focus on the country’s digital transformation and other internet-related business.9 The government had not yet announced relevant organizational regulations by the end of the coverage period.

B Limits on Content

B1 1.00-6.00 pts0-6 pts
Does the state block or filter, or compel service providers to block or filter, internet content, particularly material that is protected by international human rights standards? 6.006 6.006

The government does not generally compel service providers to block or filter websites or social media platforms. However, certain laws authorize the restriction of content online (see B3).

Although the services remain available and unblocked for users looking to access them online, the government increased its efforts to restrict content on Chinese streaming video platform iQIYI, which is owned by the Chinese firm Baidu, during the coverage period.1 In August 2020, the Ministry of Economics announced that starting in September Taiwanese companies could not provide video streaming-related services originating with Chinese companies or people, particularly iQIYI or Tencent. The rule updated the Act Governing Relations between the People of the Taiwan Area and the Mainland Area, and formally prohibited companies and individuals in Taiwan to be the agent or distributor of any Chinese over-the-top services (OTT) via television or other broadcast, including the digital-television channel service Media on Demand.2

The Taiwan Association of Human Rights (TAHR) reported in 2018 that the Taipei city government filtered certain content on its free Wi-Fi services provided to public spaces.3 For example, the city government confirmed that it filtered websites related to drug abuse, adult content, gambling, phishing, sex education, and weapons. Information agencies in the cities of New Taipei, Taichung, Tainan, and Kaohsiung reported to the organization that they did not block websites on their wireless networks.

The Ministry of Education’s Network Guardian Angels (NGA) is a content-filtering software available to the public, geared toward parents and educational institutions. According to a national report, NGA was downloaded nearly 99,000 times between January 2020 to November 2020.4 The Taiwan Association for Human Rights found that the NGA’s filtered content is based on unclear standards and has targeted civil society websites, including the Taiwan Alliance to End the Death Penalty and Taiwan Tongzhi Hotline Association, a group serving the LGBT+ community.5

B2 1.00-4.00 pts0-4 pts
Do state or nonstate actors employ legal, administrative, or other means to force publishers, content hosts, or digital platforms to delete content, particularly material that is protected by international human rights standards? 3.003 4.004

Expression protected by international human rights standards is generally not forcibly removed, and intermediaries do not face onerous liability for content generated by third parties. However, a range of laws prohibit the publishing of certain kinds of content, and have permitted content removal (see B3).1 TAHR reported, for example, that the government cited the Act Governing Food Safety and Sanitation 153 times in requests to remove content between 2015 and 2016. The Copyright Act also lays out a notice and takedown procedure that obligates intermediaries to remove third party’s content that infringes on copyright.2

The judiciary has addressed cases that include requests to remove content in recent years. In February 2021, a court ruled that a defendant should remove his review of a medical clinic on Google that included inaccurate information.3 Previously in 2019, a court ruled that a former employee did not have to remove his comment about his former employer after the company claimed it included false information.4

The Institute of Watch Internet Network (iWIN) was established under Article 46 of the Protection of Children and Youths Welfare and Rights Act. The act requires that content hosts limit receiving and browsing of content deemed harmful to the physical and mental health of children and youth, such as that featuring violence, blood, sex, obscenity, and gambling.5 Among other measures, iWIN identifies this content through a complaint mechanism for users, content-screening software, promotion and review of content, a content rating system, and a self-discipline mechanism for service providers.6

iWin reported that in 2020 it received a total of 3,878 complaints: 1,391 cases related to pornography (36 percent), 1,150 cases related to content considered harmful to children (30 percent), 409 cases related to other content considered harmful to the physical and mental health of children (11 percent), 345 cases related to cyberbullying (9 percent), and 290 cases related to false information (7 percent). Among the 3,878 complaints, iWin reported 1,184 to companies, and blacklisted 1,334 pieces of content through filtering software.7 It is unclear what percentage of complaints and reports to companies led to content being removed.

In March 2021, police forced the operator of SWAG, a Taiwanese pornography platform whose servers are located in the United States, to temporarily shut down the site. The shutdown occurred after police conducted a search of the platform’s offices as part of an apparent investigation into whether it was facilitating illegal forms of sex work and offended morality as outlined in the Criminal Code.8

Google reported complying with 59 percent of 19 government requests to remove content filed from July to December 2020. Eleven requests related to electoral law, four to regulated goods and services, three to defamation, and one to copyright violations.9 Facebook restricted nine items during the same period, eight of which related to violations of local laws around international marriage services and one to violating local privacy law.10

Tech platforms have also restricted content for reasons other than requests of the Taiwanese government. A Citizen Lab report released in August 2021 found that keyword filtering attached to Apple’s product engraving service (in which users can have personal messages etched onto their Apple devices) was limiting 338 keywords.11 Restrictions included “social content” keywords such as those deemed sexually explicit, referencing illicit goods and services, and vulgarity, but also included 29 “political content” keywords, including names of high-ranking members of the Chinese Communist Party, Mao Zedong, and the Falun Gong spiritual movement.

B3 1.00-4.00 pts0-4 pts
Do restrictions on the internet and digital content lack transparency, proportionality to the stated aims, or an independent appeals process? 4.004 4.004

Online censorship is not routine in Taiwan, and government-ordered restrictions on content are grounded in law. However, civil society has raised concerns over a lack of transparency about and oversight over which government and law enforcement agencies order removal requests and how frequently they are complied with (see B2).1

A range of laws prohibit publishing certain kinds of content, including the Protection of Children and Youths Welfare and Rights Act, the Act Governing Food Safety and Sanitation, the Pharmaceutical Affairs Act, the Consumer Protection Act, and the Cosmetic Hygiene and Safety Act.2 The Statute for Prevention and Control of Infectious Animal Diseases, for example, allows the government to compel providers to block access to websites or remove webpages that sell animal products that are banned or subjected to quarantine.3

The judiciary has issued rulings around online censorship. In 2017, the Constitutional Court ruled that Article 24(2) and Article 30(1) of the Cosmetic Hygiene and Safety Act (then known as the Statute for Control of Cosmetic Hygiene), which required that manufacturers get approval from state officials before publicizing online cosmetic advertisements, were unconstitutional.4 The court validated the importance of commercial expression that helps consumers make economic choices.

The right to be forgotten was litigated during the coverage period, as part of an ongoing case. In 2018, the High Court had heard an earlier ruling by a district court in a case where the former owner of a professional baseball team requested, under the Personal Data Protection Act (PDPA), that Google remove content claiming that he engaged in illegal betting and fraud (see C6).5 The courts originally ruled that the PDPA does not explicitly protect the right to be forgotten, and that removal can only occur when the personal data is incorrect, is unlawfully processed or collected, or when the purpose for the data no longer exists. However, in February 2021, the Supreme Court ordered the High Court to hear the case again and determine whether the request is within the scope of the PDPA.6

Several bills were pending during the coverage period that relate to online content. The Digital Communication and Broadcasting Act (DCBA)—a version of which was originally published in January 2017 yet failed to pass7 —was published by the NCC during the coverage period.8 The draft imposes varying degrees of obligation on digital communication and broadcasting service providers. For instance, the draft mandates that companies disclose certain provisions in their term of services, such as privacy and security policies and channels to report inappropriate content. It also requires intermediaries to implement a notice-and-takedown mechanism to remove illegal content immediately after being notified of it in a compliant. The draft has been criticized for offloading to companies responsibility for determining content illegality.9

The Internet Audiovisual Service Management Act (IASMA), introduced in July 2020, would regulate over-the-top platforms (OTT).10 The bill was influenced by concerns that Chinese OTT services, such as iQIYI, were operating in the country without approval from the NCC in accordance with the Act Governing Relations Between the People of the Taiwan Area and the Mainland Area. The act would impose certain transparency requirements on OTT services, particularly requiring that a privacy protection policy, a cybersecurity policy, information on use of data, and avenues for user reporting of issues are disclosed in companies’ terms of service. Operators would be required to ensure that content on their platforms does not endanger national security, public order, or the moral good, or impair the emotional or physical well-being of youth. The determination of what constitutes national security is made by the relevant administrative agency. The draft also contains certain registration and reporting requirements by which companies should abide (see B6).11

B4 1.00-4.00 pts0-4 pts
Do online journalists, commentators, and ordinary users practice self-censorship? 3.003 4.004

Journalists, civil society groups, activists, and ordinary users generally do not self-censor online. However, some laws that include liability for online content—such as the Social Order Maintenance Act and criminal defamation provisions—may influence self-censorship (see C2 and C3). Self-censorship is also driven by fear of professional or legal reprisals in China and Hong Kong.

High-profile prosecutions have left some Taiwanese people who need to travel to China wary of discussing China-related issues online. For example, Taiwanese activist Lee Ming-Che was arrested by the Chinese government in 2017 while transiting through Macau and later convicted to five years in prison for "subverting state power;" social media content he posted while in Taiwan were used as evidence in court.1 Hong Kong’s National Security Law, which was implemented in June 2020, may also encourage self-censorship of China-related speech because the scope of penalties extends to speech outside of China.2 Separately, some companies, journalists, and users have issued apologies for referring to Taiwan as a country, after receiving backlash from the Chinese government and progovernment actors.3

B5 1.00-4.00 pts0-4 pts
Are online sources of information controlled or manipulated by the government or other powerful actors to advance a particular political interest? 2.002 4.004

The government does not issue formal directives or attempt to coerce online outlets to influence their reporting. However, political disinformation and online influence operations are a significant issue, particularly those which support the Chinese government’s positions or that emanate directly from Chinese party-state actors.1 The government has taken innovative action to counteract false and manipulated information in the country (see B7).

Among countries categorized as liberal democracies, the think tank Varieties of Democracy (V-Dem) in a 2019 report identified Taiwan as being one of the two most targeted by the spread of false information by foreign governments.2 Popular topics have included reunification between the Chinese mainland and Taiwan, flaws in the country’s democracy, information discrediting the government’s response to the COVID-19 pandemic, and content aimed at smearing Democratic Progressive Party (DPP) candidates, particularly during elections.3

An October 2020 report from researchers at the Taiwanese civil society group Doublethink Lab identified several disinformation tactics used to support commentary that aligns with the Chinese party-state’s positions; these originate with a range of actors including the Chinese government, the Communist Party, military agencies, private companies, and ordinary users. The tactics include financial incentives for Taiwanese outlets to broadcast pro-China narratives, the use of “content farm” websites that push out low-quality articles designed to spread quickly on social media, and the deployment of local Taiwanese online influencers and nationalist Chinese netizens to spread pro-Beijing messages.4 Another group, the US-based cybersecurity firm Recorded Future, reported in 2020 that provincial authorities in China have recruited pro-unification influencers in Taiwan with salaries ranging from $740 to $1,460 per month.5

Reports have also alleged that news outlets in Taiwan have received direction or payment from Beijing. Reuters reported in August 2019 that Chinese mainland authorities paid at least five media groups in Taiwan for coverage.6 According to the agency, China’s Taiwan Affairs Office paid RMB 30,000 ($4,600) for two favorable stories about Beijing’s attempts to attract Taiwanese businesspeople to China, which were placed in outlets Reuters did not disclose. Taiwan’s National Security Bureau also alleged in May 2019 that the Chinese government was involved in reviewing editorial content for certain Taiwanese news outlets, but declined to say which.7

A joint investigation from Doublethink Lab and the Australian Strategic Policy Institute released in August 2021 analyzed content-farm websites that target Taiwanese audiences.8 Content on one site, Qiqis.org, was found to demonstrate high bias toward Chinese-government-preferred narratives, running critical stories about the US government generally and the January 6, 2021, Capitol building attack in particular.

Two leading political parties in the country—the DPP and Kuomintang (KMT)—have each claimed that members of the other party have hired or deployed commentators to spread manipulated information online.9 In a report from the Oxford Internet Institute, interviewees claimed that individuals have been paid by election campaigns and political parties to spread messages online, however they did not clarify whether the content was false or particularly misleading.10

B6 1.00-3.00 pts0-3 pts
Are there economic or regulatory constraints that negatively affect users’ ability to publish content online? 3.003 3.003

Taiwanese users do not face onerous constraints on their ability to publish content online. Online or digital news outlets are not required to obtain a license in order to publish. Service providers are regulated by the Telecommunications Management Act and must provide services in a nondiscriminatory manner in terms of connection quality, price, condition, and information (see A4).1

Some regulations restrict online advertisement or investment originating from China. The Act Governing Relations between the People of the Taiwan Area and the Mainland Area requires government approval for mainland Chinese entities to directly own media properties and entities. It also bans advertisements from the Chinese Communist Party (CCP).2

The draft IASMA would require OTT services of a particular size, revenue, traffic, or market influence to register, or face fines between NT$100,000 and NT$1,000,000 (US$3,500 and US$35,000) (see B3).3 Foreign-owned services would be required to set up a local representative if they do not already have one, and report periodically to the NCC about the number of domestic subscribers, traffic and revenue, and user engagement.4 Local telecommunication companies that provide service to illegal OTT services from China can also face onerous fines.5

B7 1.00-4.00 pts0-4 pts
Does the online information landscape lack diversity and reliability? 4.004 4.004

Taiwan’s online information and digital media ecosystem reflects varied interests, experiences, communities, and languages. A range of newer online outlets contribute to this diversity. However, the media environment also suffers from political polarization and sensationalist content.1 A study from the Taiwan Media Watch Foundation found that people in Taiwan view the media environment more unreliable and less credible in 2019 than they did in 2014.2

Misinformation online and across Line, Facebook, Instagram, and the popular PTT online bulletin board can undermine people’s ability to access reliable information (see B5).3 A poll conducted by Doublethink Lab of voters during the country’s 2020 presidential and legislative elections found that 80 percent of respondents thought false information online was a “serious” threat, and 70 percent thought their preferred candidates had been targeted with said information.4 Separately, misinformation about vaccines spread after a surge of COVID-19 cases in May 2021, such as that vaccines could cause death in elderly people.5

The government, tech industry, and civil society have designed innovative tools to counteract the impact of false and misleading information in Taiwan (see B5).6 For example, Digital Minister Audrey Tang announced in 2019 that each government department had employed “meme engineers” to respond quickly to disinformation efforts with funny memes. Additionally, users on the messaging app Line can report information for fact-checking to the bot Cofacts, which was created by the country’s decentralized g0v civic tech community.7 Users can report content to the bot on the platform to receive information about its validity. Organizations like Doublethink Lab have also conducted innovative research to uncover and analyze disinformation campaigns and their impact.8

B8 1.00-6.00 pts0-6 pts
Do conditions impede users’ ability to mobilize, form communities, and campaign, particularly on political and social issues? 6.006 6.006

People in Taiwan can freely use digital platforms and online sources to debate and mobilize around social and political issues, including on social media platforms like Line and Facebook, as well as the online bulletin board PTT.

Current events tend to prompt considerable debate on social media. Among recent such activity, articles supporting a February 2021 referendum on the protection of algae reefs attracted thousands of shares and influenced public opinion on the issue.1

  • 1Wu Qinjie, @morethandee, “有一件事情 - 昆蟲擾西很誠摯地拜託大家 - 搶救大潭藻礁 只剩5天!... [ There is one thing that Insect Disturbance [Wu’s nickname] sincerely asks everyone – There are only five days left to rescue Tai Tam Algae Reef!...],” February 22, 2021, text and photos, https://www.facebook.com/morethandee/posts/3686548074762882.

C Violations of User Rights

C1 1.00-6.00 pts0-6 pts
Do the constitution or other laws fail to protect rights such as freedom of expression, access to information, and press freedom, including on the internet, and are they enforced by a judiciary that lacks independence? 5.005 6.006

Freedom of speech and freedom of the press are protected by Taiwan’s constitution.1 The government has also incorporated free expression and access-to-information protections under the International Covenant on Civil and Political Rights (ICCPR) into domestic law.2 The Freedom of Government Information Law was enacted in 2005.3

Taiwan’s judiciary is independent, and it provides considerable protection for speech (see C3). However, at least one court ruling has undermined strong free expression standards: in 2000, the Constitutional Court stated that the crime of defamation does not violate free speech protections in the constitution (see C2).4

C2 1.00-4.00 pts0-4 pts
Are there laws that assign criminal penalties or civil liability for online activities, particularly those that are protected under international human rights standards? 2.002 4.004

A range of laws criminalize online activities. Defamation and slander are criminal offenses. Article 309 of the criminal code imposes up to two months of detention or a fine of NT$9,000 (US$315) for publicly insulting another person, while Article 140 outlines liability of up to six months in prison or a fine if an individual “offers an insult to a public official during the legal discharge of his duties.” Some legislators have argued that Article 140 violates free expression protections and called for amending the criminal code.1

Article 310 of the criminal code imposes up to two years in prison or a fine if an individual is found guilty of “point[ing] out or disseminat[ing] a fact which will injure the reputation of another for purpose that it be communicated to the public” in writing.2 People who allege they are slandered can also request financial compensation. For defamation cases, the law excludes speech that can be proven to be true, that which is related to public concern, and that which is a “fair comment on a fact subject to public criticism.”

Several laws impose liability for disseminating false or misleading information. Under the Social Order Maintenance Act (SOMA), users can be penalized for "spreading rumors in a way that is sufficient to undermine public order and peace" with up to three days of detention or a fine of no more than NT$30,000 (US$1,050).3 The law has been used to investigate online activities (see C3).

From January 2020 to June 2021, Article 14 of the Special Act for Prevention, Relief, and Revitalization Measures for Severe Pneumonia with Novel Pathogens was in force. The article imposed up to three years of imprisonment and high fines for the dissemination of rumors or false information regarding epidemics deemed to cause damage to the public and others.4 Similarly, Article 63 of the Communicable Disease Control Act, promulgated in June 2019, outlines a fine of no more than NT$3,000,000 (US$105,000) for spreading rumors or false information about an epidemic that cause substantial harm to the public or others.5

Spreading false information during election periods can also lead to criminal penalties. Article 104 of the Civil Servants Election and Recall Act imposes a maximum of five years in prison for causing damages to the public by disseminating rumors or fraudulent content in order to elect or not elect a candidate, or for a political proposal.6 In December 2019, the legislature passed the Anti-Infiltration Act, which includes criminal penalties for spreading election-related disinformation that is instructed, funded, or sponsored by hostile foreign forces.7 After the passage of the act, pro-Beijing online media outlet Master Chain announced that it was ending operations in Taiwan.8

Under the country’s Disaster Prevention and Rescue Law, anyone who knowingly reports false information about a disaster faces fines from NT$300,000 to NT$500,000 (US$10,500 to US$17,500).9 The Food Administration Act states that no one shall “deliberately disseminate rumors or false information” relating to market food prices and the implementation of food productive programs, among other issues.10

C3 1.00-6.00 pts0-6 pts
Are individuals penalized for online activities, particularly those that are protected under international human rights standards? 4.004 6.006

Internet users in Taiwan have been investigated or prosecuted for their online activities, although cases rarely lead to significant penalties like prison terms or steep fines.

Cases under SOMA have increased in recent years, with 151 in 2019 and 233 between January and May 2020 alone.1 While the majority of SOMA cases do not lead to convictions,2 at least four from the coverage period resulted in fines. In May 2021, a user was fined NT$5,000 (US$175) for sharing inaccurate information in a Facebook group claiming that New Taipei City had been locked down due to the coronavirus.3 Another user was fined NT$5,000 over a March 2021 post in a LINE group that incorrectly claimed that people were being given two additional weeks of vacation due to the pandemic.4 Moreover, in April 2021, a high school student was fined NT$2,000 ($US70) for sharing on Facebook inaccurate information that the start date of the next semester was postponed to March 1 instead of the accurate date of February 25.5 Also in April 2021, another user was fined NT$6,000 (US$210) after he posted on LINE falsely claiming that the government was providing a financial subsidy to families due to the pandemic.6

Other investigations under SOMA were dismissed by the judiciary. In February 2021, well-known transgender internet celebrity Chiang Chia-wen, who uses the nickname Wang Yao claimed to be pregnant on Facebook, leading to her being investigated under SOMA.7 The Ministry of Health and Welfare claimed that Wang Yao was misleading the public and that no hospital was approved to conduct uterine transplantation. In May 2021, the court ruled that Wang Yao did not violate SOMA.8

In September 2020, a music teacher was investigated under SOMA for a Facebook post claiming that lunch provided at a Tsai cabinet meeting was extravagant and an abuse of public funds.9 Six other users were also investigated and prosecuted under SOMA for sharing the content. In October, a court ruled that although the users deliberately spread information, the content did not violate the law because it would not cause fear or panic among the public.10 Following the ruling, the administrative agency stated that it would continue to investigate and prosecute rumors. During the previous coverage period, in December 2019, National Taiwan University professor Su Hongda was questioned for violating SOMA over a 2018 Facebook video in which he criticized the National Palace Museum's policies. In January 2020, the court ruled that the content was protected under domestic free expression provisions.11

Following the 2018 suicide of a diplomat stationed in Japan, influencer Slow Yang was charged for violating SOMA and insulting a public office. The charges came in connection with claims that she incited cyber troops to spread rumors that allegedly contributed to the diplomat’s death and insulting public officials.12 The case remained pending as of August 2020.

In mid-2020, a user was charged under SOMA for a YouTube video accusing President Tsai Ing-wen of electoral fraud during the 2020 presidential election. The court ruled in May 2020 that it was a reasonable supervision of election affairs and dismissed the case. Following the court ruling, law enforcement charged at least one user under Article 140 of the Criminal Code for insulting public office, in connection with the video.13

At least two users were also found guilty of and fined for violating Article 14 of the Special Act for Prevention, Relief, and Revitalization Measures for Severe Pneumonia with Novel Pathogens during the coverage period. One user, for example, was found to have intentionally posted to Facebook incorrect information that someone tested positive for COVID-19.14

News outlets and journalists have also been subject to criminal defamation charges. For example, in July 2019, the Want Want China Times Media Group filed a criminal defamation suit against the Financial Times’ Kathrin Hille and Taiwan’s state-owned Central News Agency (CNA).15 The charges came after Hille published an article accusing the media group of receiving regular orders from the Chinese government about their news coverage, and after CNA quoted the story in its own coverage.16 In April 2019, the media group also threatened to sue Apple Daily for its coverage claiming the group had been subsidized by Chinese government authorities.17 Separately, Apple Daily was sued for defamation because of their reporting on the personnel struggles and issues with financial reporting at Taiwan's Tatong Company.18 However, the court found the outlet not guilty in April 2021.

C4 1.00-4.00 pts0-4 pts
Does the government place restrictions on anonymous communication or encryption? 3.003 4.004

There are some limits on anonymous communication, as Taiwan has mandatory SIM card registration requirements.1 Telecommunications-related laws and regulations require service providers to record basic user information, including names and identification number, when selling all telecommunications numbers (including prepaid SIM cards).2 The NCC emphasized in 2017 that registration assists relevant agencies in criminal and fraud investigation and prevention.3

Residents of Taiwan can freely use encryption technology. The Communication Security and Surveillance Act (CSSA) authorizes law enforcement agencies to intercept wired and wireless telecommunications signals with court authorization.4 There is currently no explicit legal obligation for telecommunications companies to decrypt messages or provide decryption keys to law enforcement agencies, although they should ensure that software is compatible with interception efforts so that they can assist government surveillance.5 Some within law enforcement agencies have complained that failure to decrypt undermines criminal investigations.6

In September 2020, the Ministry of Justice released the Technology Investigation Act, which would empower law enforcement agencies that have a court order to access users’ electronic devices, including encrypted communications, via in-person contact, network transfer, or other necessary means such as malware (see C5).7

C5 1.00-6.00 pts0-6 pts
Does state surveillance of internet activities infringe on users’ right to privacy? 3.003 6.006

Taiwan’s constitution expressly guarantees secret communications and requires oversight for law enforcement agencies to monitor people’s communications.1 Judicial interpretations of the constitution have also protected the right to privacy and the right to self-determination of information.2 Additionally, the Personal Data Protection Act stipulates the collection, processing, and utilization of personal data by government agencies and the private sector (see C6).3 However, certain surveillance laws and procedures undermine these privacy rights in practice.

The Communication Security and Surveillance Act (CSSA) stipulates that an “interception warrant” approved by a court is required to access the content of communications for a range of alleged crimes that impose a minimum of a three-year prison term.4 For the same type of crimes, a prosecutor can apply for an “access warrant” from a court to access metadata records. However, in urgent situations and for specific felonies, the prosecutor does not require the court’s permission and can instead inform the enforcement authority to start surveillance.5 Within 24 hours of doing so, the prosecutor must apply for the warrant; if the court does not issue a warrant within 48 hours, the surveillance ceases. For certain serious crimes, including those that could result in prison terms of at least 10 years, prosecutors can directly access metadata without applying for a judicial warrant.6 The Code of Criminal Procedure also lays out provisions for law enforcement authorities to access personal data.7

The CSSA requires that the enforcement unit and the supervisory unit publish statistical reports about communication surveillance and communication record retrieval.8 According to the statistical report of the Judicial Yuan and the Ministry of Justice, there were approximately 56,000 communications surveillance cases9 and approximately 123,000 communication records retrieval cases in 2020.10 More than 90 percent of cases did not require the approval from a court.11 In its Internet Transparency Report, the Taiwan Association for Human Rights reported that the lack of judicial review over requests has been increasingly normalized.12

The CCSA empowers the National Security Bureau (NSB) to issue an interception warrant itself without judicial oversight during times of emergency to conduct surveillance on the domestic communication of “foreign forces or hostile foreign forces” for the purposes of national security.13 The NSB is not required to disclose its surveillance activity.

The Code of Criminal Procedure also lays out provisions for law enforcement authorities to access non-telecommunication personal data with a “search warrant” approved by a court or by receiving voluntary consent of the person being searched.14

The draft Technology Investigation Act, introduced during the coverage period, would increase authorities’ ability to monitor communications.15 For example, prosecutors could use GPS or other location-tracking tools for a two-month period of investigation without a warrant.16 The draft also authorizes police to use drone or aerial devices, on which Dirtbox-like devices—powerful devices that can facilitate surveillance by impersonating a cell phone tower—may be installed, and which may conduct surveillance for up to 30 days. Electronic devices could also be hacked into, and authorities may install malware to monitor communications (see C4). The Department of Justice cited new forms of digital crimes, particularly those coordinated on messaging apps, as necessitating the new powers.17 Civil society and other stakeholders criticized the draft’s provisions as permitting major violations to the right to privacy, and other human rights.18 It remained pending at the end of the coverage period.

Law enforcement have access to and deploy "M-Car” devices, which are base stations within a car that can intercept a target’s mobile phone signal to detect their location. After obtaining the communication records and user information, law enforcement agencies can use the M-Car device to capture International Mobile Equipment Identities (IMEIs) and International Mobile Subscriber Identities (IMSIs), and compare the different signal strengths to accurately locate the user. There are currently no clear rules for the use of M-Car devices. However, a court in February 2021 held that the use of the device to obtain location information of the parties is a legal method for law enforcement.19

It is unclear whether the government has access to spyware technology, although some reports suggest that it does. In a 2015 report, the Toronto-based organization Citizen Lab identified the Taiwanese government or law enforcement as being “suspected customers” of FinFisher, and traced FinFisher servers to the country.20 Previously, government agencies were found to have been in conversation with the now defunct Italian firm Hacking Team about purchasing spyware, although there is no evidence that it was purchased.21

There are also concerns that state agencies conduct social media surveillance. The NSB admitted in 2018 that they monitor social media in order to track disinformation emanating from China and to ensure national security.22 Other government units have also been found to have purchased monitoring and analytic system.23

C6 1.00-6.00 pts0-6 pts
Does monitoring and collection of user data by service providers and other technology companies infringe on users’ right to privacy? 3.003 6.006

The Personal Data Protection Act (PDPA) governs the collection, processing, and usage of personal data, including by the private sector and nongovernmental agencies. The law broadly defines personal data to include any data that can be used to directly or indirectly identify an individual, including medical information, education, financial situations, social activities, and more. PDPA also regulates the cross-border transfer of data,1 and stipulates that individuals can apply to a court for relief if a public or private actor violates the law.

The PDPA lacks an independent and dedicated competent authority overseeing its implementation, although the law’s Enforcement Rules include regulation and supervision in a more decentralized manner. Since 2018, for example, the National Development Council has established a dedicated office overseeing PDPA.2 In 2019, the government solicited public comments to determine whether the PDPA should be amended to fully comply with principles in the European Union’s General Data Protection Regulation.3

The TMA and the CSSA require service providers and the telecommunications industry to cooperate with criminal investigations and comply with law enforcement and other government authorities’ surveillance requirements (see C5).4

Government units with certain investigative powers have also gone directly to state agencies and private companies to request personal data without first receiving a court order or other oversight.5 For example, the Ministry of Economic Affairs between 2017 and 2018 had a 100 percent success rate in receiving information from the 1,112 requests it filed for personal information. 112 of the requests were to government agencies, with 1,000 to nongovernment agencies, including Chunghwa Telecom, Taiwan Mobile CO., and Yahoo! Taiwan Holdings Limited.6 Between 2015 and 2016, the Ministry of Finance submitted 350 requests with a 99.4 percent success rate. The Criminal Investigation Bureau also reportedly issued 565 requests to Facebook through this process, with a 52.9 percent success rate, between 2015 and 2016.7

Several laws mandate different requirements for data retention.8 Telecommunication providers are required to store communication records, subscriber information, and billing details for at least a year.9

The government increased data collection and other monitoring during the COVID-19 pandemic, practices that have been criticized by civil society groups and other experts as lacking legality and proportionality.10 In February 2020, the Special Act for Prevention, Relief, and Revitalization Measures for Severe Pneumonia with Novel Pathogens was enacted, giving the Central Epidemic Command Center (CECC) broad power to conduct contact tracing and publicize personal information.11 The act states that personal information will be “processed in accordance with related regulations for personal data protection after the end of the pandemic” and the government has also claimed it will delete stored data after the pandemic.12

The Electronic Fence System, which uses mobile location tracking data to ensure individuals remain in quarantine, was in effect during the coverage period.13 The CECC can access aggregated data from the system, and police responding to quarantine-related alerts can access an individual’s name, phone number, and address. Those in quarantine must keep their phones on in order for the tracking to work.

In May 2021, the government introduced the 1922 SMS contact-tracing system, which uses QR codes to track when users enter or leave different locations, including stores, government buildings, and public transportation venues.14 Mobile providers help facilitate the tracking: once an individual tests positive for COVID-19, a contact tracer provides their phone number to the telecom provider and then the company analyzes location data to determine who was in proximity of that individual. The data is purportedly only supposed to be used for epidemiological investigations.15 However, in June, a judge of the Taichung District Court alleged that the service was used to locate an accused individual in a criminal investigation.16 The NCC responded by again stating that the text service is only used for epidemiological purposes.17

Additionally, in February 2020, a collective infection on the cruise ship Diamond Princess prompted the government to obtain mobile phone location information of over 600,000 people from telecommunication companies, for contact tracing.18

C7 1.00-5.00 pts0-5 pts
Are individuals subject to extralegal intimidation or physical violence by state authorities or any other actor in relation to their online activities? 4.004 5.005

Users are generally free from physical violence or other serious threats due to their online activity, although online harassment remains a concern.

“Cyber manhunts” refer to the identification and tracking down of someone following criticism or their involvement in controversial events, and often include doxing. In early April 2021, following a fatal derailing of a train, online users tried to identify who caused the crash. One passenger was mistakenly accused of being at fault and was subject to online harassment.1 In December 2020, after an airline pilot was infected with COVID-19 and broke Taiwan’s 253-day zero-diagnosed record, the pilot’s personal information and family background were exposed online by users.2 Another woman was incorrectly alleged to have contracted the virus from the pilot.

Brian Hioe, editor for the New Bloom Magazine, reported facing harassment and doxing after he was quoted raising privacy concerns about Taiwan's COVID-19 measures tracking people in quarantine in a BBC article.3

Although not routine, users have faced physical threats or violent incidents in relation to online activities during previous coverage periods. In March 2020, prominent YouTube personality Tsai Aga and his wife were attacked by an unknown assailant in connection with his videos.4

LGBT+ users have experienced online harassment.5 Separately, in April 2021, the Executive Yuan passed a draft of Anti-Stalking Act that is in part expected to deal with online sexual harassment.6

C8 1.00-3.00 pts0-3 pts
Are websites, governmental and private entities, service providers, or individual users subject to widespread hacking and other forms of cyberattack? 1.001 3.003

Taiwan faces frequent overseas cyberattacks, emanating from Beijing in particular.

The country’s Department of Cyber Security said in 2019 that Taiwan faced about 30 million technical attacks every month, such as webpage defacements and DDoS attacks, half of which are speculated to originate in China.1 In September 2020, the government reported that at least 10 agencies and an estimated 6,000 email accounts had been targeted since 2018,2 with four Chinese government-backed hacking groups believing to have been involved.3

In its 2020 country report on human rights practices, the US Department of State noted that Chinese government actors have also conducted cyberattacks against Taiwanese journalists’ computers and mobile phones.4

In May 2020, around the inauguration of the president, documents from the staff of the presidential office were leaked to reporters.5 Government officials also claimed certain documents were altered. Also in May, hackers sent phishing emails from the Office of the President to legislators.6 There have been no reports or further information about the identity of the hackers.

Data leaks are also an issue in Taiwan. In May 2020, an international security company pointed out that there are more than 20 million Taiwanese household registration data items circulating on the dark web.7

Taiwan’s Cyber Security Management Act oversees the cybersecurity of critical infrastructure providers. It requires that public agencies formulate cybersecurity maintenance plans as well as stipulates reporting and responding mechanisms when facing security incidents.8 The Executive Yuan also established a Cyber Security Office.9

On Taiwan

See all data, scores & information on this country or territory.

See More

Country Facts

  • Global Freedom Score

    94 100 free

  • Internet Freedom Score

    78 100 free

  • Freedom in the World Status

    Free

  • Networks Restricted

    No

  • Websites Blocked

    No

  • Pro-government Commentators

    No

  • Users Arrested

    Yes