4. Media freedoms and civic space in the digital age for transparency, accountability and citizen participation

Abstract

This chapter provides an overview of the status of press freedom and civic space in a digitalised world, including relevant legal frameworks. It discusses harassment and attacks targeting journalists and makes suggestions on building the necessary enabling environment for reliable, fact-based journalism. It considers the protection of online civic space for citizens and related challenges such as hate speech and mis- and disinformation. It concludes with an analysis of the importance of personal data protection for civic space and safeguarding civic freedoms in the context of increased use of artificial intelligence (AI).

Key findings

Protected media freedoms are an essential component of democratic societies. Freedom of the press is guaranteed in law in all respondents but the level of harassment and threats targeting journalists, including those covering protests, remains a serious concern.
In recent years, a number of OECD Members and non-Members have passed special measures enhancing the rights of journalists or protecting them against threats of violence or intimidation. Courts have also issued decisions strengthening the rights of journalists and media. However, in some countries, national security measures have the potential to stifle freedom of the press.
While the overall picture for freedom of the press is good in OECD Members compared to the rest of the world, it has deteriorated in recent years. The proportion of countries where the situation is regarded as favourable for journalism has halved in the space of six years, with 49% of countries ranked as “good” in the 2015 World Press Freedom Index and only 26% in 2021, according to data from Reporters Without Borders (RSF). The level of press freedom afforded to journalists was classified as “good” or “fairly good” in nearly three-quarters (74%) of all OECD Members in 2021 and “problematic” or “bad” in the rest (26%).
The protection of online civic freedoms is a precondition for citizens and civil society to access information, operate freely online and thrive without fear of arbitrary or unlawful intrusion. Almost all respondent OECD Members (94%, 75% of all respondents) have legal provisions to protect an open Internet and a number have passed recent legislation and other measures to enhance access to and safety of the Internet and net neutrality.
Nevertheless, in 93% of respondent OECD Members (86% of all respondents), the Internet can be restricted to protect national security and measures introduced in a few countries have intensified pressure on content-sharing services to filter illegal content. Depending on how these are implemented, such measures may risk stifling public debate by overly restricting online content.
Online hate speech and harassment are a growing obstacle to online civic participation, with countries introducing a variety of measures to tackle the phenomenon: 65% of respondent OECD Members have established reporting and complaint mechanisms and provide support for victims (e.g. hotlines, free legal advice) and almost half (44%) have introduced specific measures to address online hate speech that targets women.
Concerns about the misuse of artificial intelligence (AI) are also growing and have particular prominence in the use of AI in and by the public sector. Almost one-third of the 19 reviewed strategies include an in-depth discussion on the impact of AI on civic freedoms. More than half of these strategies address the need to establish oversight and redress mechanisms and a majority (84%) include the intention of developing an ethics framework to guide the development of AI, in particular in the public sector.

4.1. Introduction

Protected media freedoms are an essential component of democratic societies. They allow for access to diverse sources of information and enable informed debate as part of a vibrant “public interest information system” (Forum on Information and Democracy, 2021[1]) that facilitates citizen and stakeholder participation in public life and decision making. Media restrictions, including media concentration and monopolies, in contrast, can hamper balanced and multifaceted debate on matters of public interest and promote one-sided views that can ignite polarisation, in addition to impeding transparency and accountability.

Online civic freedoms are an equally crucial component of an information ecosystem in democratic societies, understood for the purposes of this report as the combination of communication and media governance frameworks (i.e. institutional, legal, policy and regulatory) as well as principal actors (i.e. governments, traditional and social media, professional and citizen journalists) (Matasick, Alfonsi and Bellantoni, 2020[2]). Citizens 1 and civil society organisations (CSOs) are increasingly moving their activities onto social media and the Internet as part of what is commonly referred to as online civic space. This shift has been accentuated by the COVID-19 pandemic, related lockdowns and restrictions on freedom of assembly due to public health measures. At a time when civic space – defined as the set of legal, policy, institutional and practical conditions necessary for non-governmental actors to access information, express themselves, associate, organise and participate in public life 2 – is eroding globally, digital transformation is thus providing new opportunities to exercise key rights, support civic mobilisation and the articulation of interests, and facilitate more dynamic and inclusive civic participation (Freelon, 2016[3]; Anderson, 2018[4]).

Technological advances and social media platforms are powerful tools for civic engagement, yet data-intensive technologies also come with important challenges to civic freedoms and democratic governance. In some countries, digital technologies are being misused or abused by governments to surveil or even silence civil society and stifle political opposition, as well as to express extremist views and hate speech, undermining the safety and security of online civic space (OECD, 2020[5]). More and more citizens and civil society actors are thus demanding transparent and accountable governance, regulations and processes for online civic spaces that are open, accessible, safe and equitable and that serve the public interest.

Online civic space is also becoming more complex as a result of the increasing role played by online platforms and of the rapidly advancing technologies such as AI and facial recognition, impacting civic freedoms, including privacy and raising additional concerns about their ethical use. Citizens are often unaware of how their online statements are turned into data, how systems used by online platforms sort content and how they profile and target them through advertising (Zuboff, 2019[6]; Couldry and Mejias, 2019[7]). Governments also recognise technological advances and the (mis)use of personal data as growing challenges to civic freedoms. According to the 2019 OECD Digital Economy Policy Questionnaire, conducted under the purview of the OECD Committee on Digital Economy Policy (CDEP), more than 80% of the 29 OECD respondents reported in 2019 that AI and big data analytics are currently the biggest challenges to privacy and personal data protection (OECD, 2020[8]). Similarly, as addressed by the CDEP initiative on terrorist and violent extremist content (TVEC), limited reporting by platforms of their TVEC moderation is hampering constructive dialogue and an understanding of the impacts on online civic space.

All survey data presented in this chapter pertain to the countries that responded to the civic space section (32 OECD Members and 19 non-Members) of the 2020 OECD Survey on Open Government (hereafter, “the Survey”). The chapter also benefits from content contributed by RSF, which covers all 38 OECD Members.

4.2. Protection of freedom of the press

4.2.1. Legal frameworks governing freedom of the press

Media freedom is a fundamental component of civic space as a prerequisite for the unrestricted flow of information and the open exchange of opinions and ideas. The right to freedom of expression, discussed in detail in Section 2.1.1 of Chapter 2, underpins media freedom in national legislation and the principles set out there apply equally to media freedoms. According to the United Nations (UN) Human Rights Committee, countries are obliged to ensure that legislative and administrative frameworks for the regulation of mass media, including in print, broadcast and online media, are consistent with the provisions of Article 19 (para. 3) of the International Covenant on Civil and Political Rights (ICCPR), which limits the circumstances in which the right to freedom of expression may be restricted (UN Human Rights Committee, 2012[9]). The UN has also affirmed that the same rights that people have off line must also be protected online, particularly freedom of expression (UN, 2021[10]).

For the purposes of this report, press freedom is defined as the principle that communication and expression through various channels, including printed and electronic media, are considered a right to be exercised freely and without interference from an overreaching state polity. The OECD Survey on Open Government shows that in all respondents 3 the principle of freedom of the press is set out in relevant constitutional and/or media-specific legislation or guaranteed by high court decisions, either explicitly or as part of a general constitutional right to freedom of expression.

In many countries, the exceptions to this right, notably legislation prohibiting and sanctioning defamation, blasphemy, hate speech, incitement to violence and insults to heads of state, are the same as exceptions to the general right to freedom of expression (Section 2.1.1 in Chapter 2) and are shown in Figure ‎4.1. A total of 92% of OECD respondents have legal exceptions to press freedom based on defamation and incitement to violence (95% of all respondents), while 85% make exceptions on the basis of hate speech (81% of all respondents). Furthermore, 54% of respondent OECD Members (56% of all respondents) exclude false advertisement/misrepresentation of facts from the right to freedom of the press. Exceptions based on other grounds are less common, such as blasphemy (27% of respondent OECD Members, 30% of all respondents) and insults to heads of state (27% of OECD Members, 28% of all respondents).

Figure ‎4.1. Legally mandated exceptions to freedom of the press, 2020

A number of OECD Members also have more specific press-related limitations in legislation covering media and communications. For example, there is a prohibition of hate speech in mass media broadcasting and advertising in Costa Rica and Latvia and a ban on publishing information that is slanderous and offensive, injures the honour and dignity of people or otherwise defames them in Latvia and Lithuania. In a number of respondents, including Austria, Brazil, Colombia, Cameroon, Germany and Spain, penalties for hate speech are increased when these are committed through media or publication by any means. Other countries, for instance, Italy, foresee aggravated punishment in cases where members of the press commit acts of defamation. In Greece, criminal legislation likewise aggravates punishment for the repeated act of spreading “false news likely to cause concern or fear among citizens or undermine public confidence in the national economy, the defence capacity of the country or public health” committed by the media; and media owners and publishers are also liable for news provided by their media outlet (CPJ, 2021[11]). Another example is Ireland, which, in its constitution, states that the press may not be used to undermine public order, while in Latvia, programmes and broadcasts issued by electronic mass media may not contain incitement to violence or hatred, or incitement to discrimination against a person or group of persons.

While it may be justified to provide aggravated sanctions for media outlets when it comes to hate speech or defamation, relevant provisions should be formulated in a clear and foreseeable manner, so that journalists know which laws protect their rights in case of violations and that the respective sanctions are not so high as to unduly limit the rights of media or individual journalists, or hamper the ability of media outlets to function.

Additionally, high courts in Germany, Lithuania and Türkiye have issued decisions strengthening the rights of journalists and media in general in cases related to state surveillance or other forms of state interference, access to information, oversight of public broadcasting and the accreditation of media representatives. These court rulings are a positive sign that in various OECD Members, the judiciary is protecting and safeguarding civic space. Box ‎4.1 highlights special measures that have been implemented to protect journalists.

Box ‎4.1. Laws, policies and programmes for the physical protection of journalists

A number of respondents to the OECD Survey on Open Government have passed special measures enhancing the rights of journalists, providing them with additional support when conducting their work, or protecting them against threats of violence or intimidation. Human rights defender protection laws in Honduras and Mexico (Section 2.3.1 in Chapter 2) explicitly apply to journalists, while Colombia has passed additional legislation and policies to protect journalists and social communicators (Government of Colombia, 2000[12]). Mexico has a special prosecutor’s office that investigates crimes against journalists (UNESCO, 2021[13]) and, in Portugal, murder is met with aggravated sanctions if committed against a journalist.

In response to the rising levels of threat to journalists, several OECD Members have also started to develop specific policies to protect them. The United Kingdom, for example, has a national action plan to protect journalists from abuse and harassment, including measures for training police officers and journalists, while social media platforms and prosecution services have committed to taking prompt and tough action against abusers (UK Government[14]). In Brazil, since 2018, the federal human rights protection programme of the Ministry of Human Rights has also explicitly protected “communicators”, defined as persons performing regular social communication activities to disseminate information aimed at promoting and defending human rights (Ministry of Human Rights, 2018[15]).

Source: Government of Colombia (2000[12]), Decree 1592 of 2000, https://www.suin-juriscol.gov.co/viewDocument.asp?id=1314526; Government of the United Kingdom (2021[14]), “Government publishes first ever national action plan to protect journalists”, https://www.gov.uk/government/news/government-publishes-first-ever-national-action-plan-to-protect-journalists; Ministry of Human Rights (2018[15]), Ordinance Nº300 of 3 September 2018, https://www.in.gov.br/materia/-/asset_publisher/Kujrw0TZC2Mb/content/id/39528373/do1-2018-09-04-portaria-n-300-de-3-de-setembro-de-2018-39528265; UNESCO (2021[13]), “Recent convictions highlight the work of Mexico’s Prosecutor Office dedicated to crimes against freedom of expression”, https://en.unesco.org/news/recent-convictions-highlight-work-mexicos-prosecutor-office-dedicated-crimes-against-freedom.

There have been numerous amendments to legal frameworks and court decisions governing media freedoms in recent years aimed at safeguarding the rights of the media and introducing clauses based on national security and counter-terrorism concerns. OECD Members such as Canada, Germany4 and the Netherlands have recently passed legislation that protects journalists and their sources from undue disclosure and surveillance measures. In Canada, the Supreme Court ruled in 2018 that a Vice reporter had to surrender his materials related to a case about an accused terrorist to the Royal Canadian Mounted Police. The court acknowledged the potential negative impact of such a decision on journalists and their secret sources but said the “state’s interest in the investigation and prosecution of crime outweighed the media’s right to privacy in gathering and disseminating the news”.5 Norway has passed a media liability act relating to editorial independence that clarifies the liability of editor-controlled journalistic media and thereby contributes to open and informed public discourse (Box ‎4.2). Italy has likewise amended its legislation on the media so that defamation is no longer punishable by cumulative penalties (involving prison sentences and fines); rather, while still a criminal offence, the sanction is now either a prison sentence or a fine.

Box ‎4.2. Good practice: Media Liability Act in Norway

In Norway, since 2020, a new media responsibility law defines journalists’ freedoms and responsibilities. The purpose of the act is to facilitate open, informed public discourse by ensuring editorial independence and establishing clear liability regulations for content published in editor-controlled journalistic media. The duties of media professionals defined in the law include the obligation to have an editor (Section 4 of the law) and for that person to be known (Section 5 of the law). The editor is responsible for the content of the medium but also for the clarity of the rules applying to user-generated content. In cases where a medium hosts editorial content and user-generated content, users should be able to distinguish these (Section 6 of the law). In addition, the law also ensures the editorial independence of the medium stating that the publisher, the owner or other company management should not interfere with the editor’s final decision on content (Section 7 of the law).

Source: (Norwegian Ministry of Culture and Equality, 2020[16]); (Reporters Without Borders, 2021b[17]).

At the same time, threats against national security, such as terrorism have led to more restrictive legislation in a number of OECD Members. In Australia, journalists may be categorised as people “acting on behalf of a foreign principal” in certain circumstances, which means that they may be required to register in a public registry and disclose foreign contacts and may suffer sanctions for failure to do so (Parliament of Australia, n.d.[18]). In Latvia and Lithuania, broadcasting or retransmission permits may be refused or suspended where necessary in the interests of national security or public order. Similarly, the Netherlands has recently passed legislation that obliges anyone, including journalists, travelling to areas controlled by terrorist groups to request prior permission from the Ministry of Justice.

Additionally, some respondents have introduced new provisions in their criminal codes or adopted new legislation in recent years to combat mis- and disinformation,6 including Greece (2021), Kazakhstan (2018), the Philippines (2017) and Türkiye (2022), with prison sentences of up to three years as potential sanctions. Human rights bodies and CSOs have raised concerns as general prohibitions on the dissemination of information are often based on overly broad and vague language or concepts (ICNL, 2021[19]; UN, 2022[20]; RSF, 2021[21]).

Generally, most OECD Members provide for some sort of complaints and redress mechanisms through which individuals and media entities can complain about alleged violations of freedom of the press. These mostly range from constitutional and other courts to government entities and independent institutions, such as national human rights institutions (NHRIs). Broadcasting media may also complain about licensing issues to independent broadcasting councils, such as those in the Czech Republic. Similarly, the United Kingdom’s regulatory and competition authority, Ofcom, investigates complaints and can revoke broadcasting licenses. In addition, in some countries, individuals can address complaints about the press or broadcast matters to specific complaint bodies. In Austria, for example, the Presserat allows for individual complaints regarding news content that violates the code of ethics established by the institution, as well as complaints where an individual is personally affected. In New Zealand, the Media Council’s complaint procedure foresees that the council rulings following a complaint are made public. In Korea, the Press Arbitration Commission allows individuals or press organisations to initiate a mediation procedure.

Key measures to consider on legal frameworks on media freedoms

Ensuring that legal frameworks guarantee media pluralism, the independence of journalists and the media, and the right of journalists to protect the secrecy of their sources; and that freedom of the press is also fully protected in the context of security or counter-terrorism measures.

4.2.2. Implementation challenges and opportunities for media freedoms, as identified by CSOs and other stakeholders

Harassment and attacks on journalists

Journalists enjoy special protection based on the right to freedom of information and expression under Article 19 of the ICCPR, Article 10 of the European Convention on Human Rights (ECHR) and Article 13 of the American Convention on Human Rights (ACHR). In 2020, the UN Human Rights Council passed a resolution on the safety of journalists, condemning related attacks, reprisals and violence, including specific attacks on women journalists and impunity for such acts (UN Human Rights Council, 2020[22]). In 2022, the Council of Europe published its principles on Journalism in Situations of Conflict and Aggression, which are based on the relevant conventions, recommendations, guidelines and case law from the European Court of Human Rights (Council of Europe, 2022[23]).

Despite these special protections, estimates of journalists killed worldwide between 2010 and 2020 range between 937 (RSF, 2020[24]) and 956 (UNESCO, 2021[25]). The COVID-19 pandemic has further exacerbated the threats and limitations to journalists’ work (OAS, 2020[26]). In 2021, 293 journalists were imprisoned worldwide, reaching a new high, according to the Committee to Protect Journalists (CPJ, 2021[27]) (Figure ‎4.2). In the same year, estimates of journalists killed range from between 31 to 55 killings (Section ‎4.3.3 on the growing vilification of journalists).7 Between 2017 and 2021, 67 journalists and media workers were killed in respondent countries with the motive confirmed as related to their work (CPJ, n.d.[28]). Some countries are experiencing particular challenges in relation to protecting journalists in contexts marked by organised crime and social conflicts (CPJ, 2021[29]; UNESCO, 2021[25]).

Figure ‎4.2. Number of journalists killed between 2017 and 2021

A related global trend is the increase in attacks against journalists covering protests. Around the world, journalists have been increasingly targeted by harassment and violence at demonstrations, with a United Nations Educational, Scientific and Cultural Organization (UNESCO) report finding 125 such instances across 65 countries between January 2015 and June 2020 (2020[31]). Journalists can face verbal and physical intimidation both from protestors and the police. In 2021, RSF reported on instances of police violence at protests negatively impacting press freedom, including journalists suffering injuries from teargas and baton strikes and others having their equipment seized (RFI, 2021[32]). The situation can be particularly difficult for independent and freelance journalists, who often face barriers in accessing events and can be more vulnerable to harassment without the protection of a larger media company.

Beyond constituting human rights violations, attacks against journalists limit free expression and deprive others of their rights to receive information, thus hampering freedom of expression, public debate and civic space more broadly. In 2021, reflecting the growing concern about violence targeting journalists in Europe, the European Commission (EC) adopted a Recommendation on the Protection, Safety and Empowerment of Journalists and Other Media Professionals in the European Union (EU) (European Commission, 2021[33]). In 2016, the Committee of Ministers of the Council of Europe adopted a Recommendation on the protection of journalism and the safety of journalists and other media actors (CoE, 2016[34]).

The V-Dem Institute’s indicator on the harassment of journalists 8 for 2020, which is based on expert evaluation, shows that in 75% of respondent OECD Members, it was rare for a journalist to be harassed for offending powerful actors and, if this were to happen, those responsible for the harassment were identified and punished (V-Dem Institute, 2021[35]). In 16% of respondent OECD Members, some journalists who offended powerful actors were forced to stop working but others managed to continue practising journalism freely. In only one OECD respondent country, journalists who occasionally offended powerful actors were almost always harassed and eventually forced to stop (V-Dem Institute, 2021[35]). In two OECD Member respondents, journalists were never harassed by governmental or powerful non-governmental actors while engaged in legitimate journalistic activities.

Key measures to consider on protecting journalists

- Establishing effective mechanisms, initiatives and programmes to protect journalists at risk and to systematically investigate and provide access to justice for threats and attacks to ensure full accountability.

- Taking additional measures to increase training among the police on protecting journalists’ ability to report and on ensuring they themselves follow protocols to avoid any escalation of violence.

- Engaging in awareness-raising on the crucial role that journalism plays in democratic societies, as part of a healthy public interest information ecosystem.

Ensuring media pluralism and avoiding capture by other interests

A plurality of media owners contributes to a more effective “watchdog” environment by reducing the risk of public opinion being dominated by a single actor. Relying solely on publicly owned media makes it difficult to gauge whether reporting is unbiased. Likewise, relying only on privately-owned media may result in media “moguls” who use their position to exert undue influence on news content. Therefore, opting for and promoting a mix of both public and private media can help ensure a balance (Stapenhurst, 2000[36]).

Similarly, transparency in media ownership is also crucial. For the public to evaluate the objectivity of specific media outlets and for the government to evaluate media diversity, the business interests of media owners should be transparent and accessible to all. Transparency on business activity between private media and governments is even more relevant to prevent any form of undue political influence. Governments could consider instituting measures to require disclosure of business interests to an independent regulator or directly to the public in the form of a publicly available registry, or both. Governments could also consider establishing transparency measures to identify the beneficial owners of the media, especially in the broadcasting sector (OECD, 2020[37]).

4.3. Freedom of the press in OECD Members: Contribution from Reporters without Borders

4.3.1. The status of press freedom in OECD Members according to RSF

The World Press Freedom Index from Reporters Without Borders (RSF), an international non-profit organisation working to defend the right to access free and reliable information, evaluates the level of media freedom in 180 countries and territories every year and is widely used as a reference for media freedom (Box ‎4.3). Its 2021 edition shows that journalism is blocked or seriously impeded in 73 countries and is restricted in 59 others, which together represent 73% of the countries ranked by RSF.

Box ‎4.3. How Reporters Without Borders’ World Press Freedom Index is compiled and the main trends of 2022

Published annually by RSF since 2002, the World Press Freedom Index measures the level of media freedom in 180 countries and territories. It assesses the level of pluralism, press independence, the environment for the media and self-censorship, the legal framework, transparency and the quality of the infrastructure that supports the production of news and information. The index does not evaluate government policy.

The global indicator and regional indicators are calculated on the basis of scores registered for each country or territory. In 2021, these scores were calculated from the answers to a questionnaire available in 20 languages, completed by experts throughout the world, supported by a qualitative analysis.

In 2021, RSF used seven indicators for its ranking: pluralism, media independence, environment and self-censorship, legislative framework, transparency, infrastructure and abuses. Countries are classified as “good”, “fairly good”, “problematic”, “bad” or “very bad”. The same weighted average method has been used to calculate the overall indicator for the 38 OECD Members.

For its 20^th edition in 2022, RSF updated its methodology, basing the Index’s rankings on a new score ranging from 0 to 100, 100 being the best possible rating. These scores are based on a tally of abuses against journalists and on the responses of press freedom specialists (journalists, researchers, academics and human rights defenders) to an RSF questionnaire available in 23 languages. Each country and territory are evaluated through five contextual indicators: political context, legal framework, economic context, sociocultural context and safety.

The main trend in 2022 is a double polarisation: i) within democratic societies, divisions are growing as a result of the spread of opinion media and of disinformation circuits that are amplified by the way social media function; and ii) within autocratic countries, propaganda and media shutdowns are more problematic than ever.

In Europe, the Nordic countries at the top of the Index – Denmark, Norway and Sweden – continue to serve as a democratic model where freedom of expression flourishes. But the region shows significant disparities and conditions on both extremes have evolved considerably. Estonia (4^th) and Lithuania (9^th) – two former communist states – are now among the top ten, while the Netherlands (28^th) no longer is. Greece (108^th) is last in Europe. These developments can partly be explained by the return of murders of journalists in the EU: Giorgios Karaivaz, in Greece, and Peter R. De Vries, in the Netherlands, were gunned down in the centre of two European cities.

Overall, European institutions have started to implement protective measures for journalists and press freedom and they have launched proceedings against Hungary (85^th) for violating European law. It is worth noting, nevertheless, that Slovenia (54^th), Poland (66^th), Albania (103^rd) and Greece also intensified draconian measures against journalists.

Source: Reporters without Borders (2021[38]), “Detailed methodology”, https://rsf.org/en/detailed-methodology.

The results for the 38 OECD Members are more positive overall (Figure ‎4.3). The level of freedom enjoyed by journalists was classified as “good” (grey) or “fairly good” (yellow) in nearly three-quarters (74%) of OECD Members in 2021, while the proportion of all 180 countries with a “good” or “fairly good” evaluation is only 27%. Similarly, the freedom of the press situation is deemed to be “problematic” (orange) or “bad” (red) in just one-quarter of OECD Members (26%), while this proportion is more than 60% in all 180 countries. Significantly, no OECD Member is designated as black, signifying a country where the situation is “very bad”.

Figure ‎4.3. Overall press freedom scores in OECD Members and across 180 countries, 2021

While the overall situation continues to be satisfactory in OECD Members compared to other countries, it has nonetheless eroded in recent years. The proportion of OECD Members where the situation is regarded as good for journalism has been halved in the space of six years. The proportion of countries ranked as “good” (grey) was 49% in the 2015 World Press Freedom Index but has fallen to 26% in 2021 (Figure ‎4.4). This decline is attributable above all to the various crises that journalism has experienced: a geopolitical crisis (due to the aggressiveness of authoritarian regimes); a technological crisis (due to a lack of democratic guarantees); a democratic crisis (due to polarisation and repressive policies); a crisis of trust (due to suspicion and even hatred of the media); and an economic crisis (that has impoverished quality journalism) (RSF, 2020[40]). All of these crises have been compounded by the coronavirus (COVID-19) crisis since early 2020.

Figure ‎4.4. Evolution of a “good” rating in press freedoms among OECD Members, 2015-21

4.3.2. RSF indicators for OECD Members

The seven indicators that RSF uses to compile the World Press Freedom Index every year – which measure factors such as the level of media independence or the climate in which journalists work and the degree to which they may feel the need to censor themselves for their protection – show better results overall for OECD Members than the rest of the world.

Figure ‎4.5. World Press Freedom Index indicators with better results across OECD Members: Independence, 2017-21

Figure ‎4.6. World Press Freedom Index indicators with better results across OECD Members: Transparency, 2017-21

Within a year, the level of violence against journalists has doubled in Europe (which includes 27 OECD Members), whereas the deterioration worldwide was 17%. The deterioration has been due above all to an increase in attacks against field reporters. Many cases of police violence were reported amid an increase in protests worldwide. Journalists were also attacked by supporters of extremist and conspiracy-theory groups during protests against restrictions imposed to combat the COVID-19 pandemic.

The Environment and Self-Censorship indicator, which evaluates the environment for journalists and the pressures to which they may be subjected, registered little variation among OECD Members between 2019 and 2021. However, the Transparency indicator, which measures restrictions on reporters’ access to information – whether in the field or from sources – registered a sudden worsening in 2020. This was directly linked to the COVID-19 pandemic and lockdown measures, which drastically restricted journalists’ reporting and coverage of events (Figure ‎4.11 on access restrictions). The Independence indicator, which measures the degree to which the media are able to function independently of sources of political, governmental, business and religious power and influence, registered a 10% improvement over 2020 within OECD Members.

The overall trend must, however, be nuanced, given a thorough analysis of the responses by media professionals and different experts to the questionnaire used to help compile the World Press Freedom Index. The likelihood of state interference in the appointments and dismissals of public media journalists and executives is considered significant in one-third of the 38 OECD Members (Figure ‎4.7). This proportion corresponds to the number of countries where the experts put the likelihood of interference at 5 out of 10 or more, with 10 out of 10 signifying complete freedom to have a journalist or executive fired from a state-owned media company or an executive fired from a privately-owned media company. This problem is particularly marked in Central and South America and the eastern part of Europe. State media executives are particularly exposed to dismissal in 19 OECD Members and privately-owned media executives are also exposed to this threat in 7 countries.

Figure ‎4.7. Public media exposure to dismissals in OECD Members, 2021

Nevertheless, while privately-owned media are resisting better overall to pressure from authorities, they are more sensitive to economic pressures. The responses to the question “To what degree are privately-owned media economically dependent on direct or indirect subsidies?” indicate that media depend on state subsidies to function in more than half of the 38 OECD Members (Figure ‎4.8). More generally, in several OECD Members, especially in Eastern Europe, privately-owned media are exposed to fiscal, commercial and legislative pressure that may take the form of a tax on advertising revenue or can manifest itself, for example, in the form of the acquisition of local media outlets by a state-controlled company.

Figure ‎4.8. Economic dependence of privately-owned media on government subsidies in OECD Members, 2021

4.3.3. Public vilification of journalists becoming common practice

Analysis of the RSF questionnaire results reveals that journalists in OECD Members are not being spared the growing climate of mistrust and even hatred of the media, often directly fomented by politicians who publicly vilify journalists, branding them, for example, as “enemies of the people”. Journalists are regularly subjected to “public vilification”, “public insults” and “hate speech” in 23 of the 38 OECD Members (Figure ‎4.9). Although criticism and hate speech targeting journalists are also increasing in countries outside the OECD, good results from countries such as Costa Rica, Luxembourg, Norway and Switzerland show that the trend is not inevitable.

Figure ‎4.9. Public vilification of journalists in OECD Members, 2021

4.3.4. Trends in the freedom to investigate

The World Press Freedom Index also enables measurement of the level of pressure placed on journalists in connection with the subjects they cover and their freedom to investigate centres of power and influence (Figure ‎4.10). Measurement of the media’s freedom to publish revelations produces the best results (no limitation whatsoever) in Norway, Sweden, Costa Rica, Denmark and Iceland (which are ranked 1^st, 3^rd, 4^th and 16^th respectively in the World Press Freedom Index) and confirms the primacy of the “Nordic model” as regards media freedom.

On the other hand, in about one-third of OECD Members, journalists are subjected to significant constraints when investigating sensitive subjects, such as the military and organised crime. The 2021 World Press Freedom Index indicates a growing difficulty overall for journalists to investigate and publish revelations about sensitive subjects. These constraints are particularly marked, however, in Eastern Europe, Asia-Pacific and Central and South America.

The latest RSF Round-up also shows that more and more journalists are being targeted due to their investigative reporting (RSF, 2020[41]). Of the 50 journalists killed worldwide in 2020, 14 were investigating stories linked to corruption or organised crime (RSF, 2020[41]).

Figure ‎4.10. Media ability to investigate centres of power in OECD Members, 2021

The COVID-19 pandemic has placed major limitations on journalistic work and especially reporting in the field. As a result of the public health crisis, or its use as a pretext, many journalists were not able to provide coverage of many events throughout the world in 2020 (Figure ‎4.11). Of the 38 OECD Members, 10 registered a decline in journalists’ ability to access physical events in 2021, 16 reported no significant change and the remaining 12 countries registered an evolution, albeit often minimal, towards greater accessibility. This was above all the case in Central and South America, Europe-Central Asia and the Middle East.

Figure ‎4.11. The extent to which journalists faced difficulty accessing events in OECD Members, 2021

The global indicator, the measure of the level of media freedom overall, has been stable during the past five years (2016-21) in the 38 OECD Members. While some OECD Members have seen their score deteriorate during this period, others have registered significant improvements. They include Portugal, whose score has improved by nearly 5 points in the past 5 years, enabling it to rise 13 places in the World Press Freedom Index. Overall, the OECD Member results continue to be better, on average, than those of all 180 countries and territories ranked by RSF (Figure ‎4.12).

Figure ‎4.12. OECD Member and world global scores for press freedom, 2016-21

The main trends of the 2022 World Press Freedom Index reflect a “two-fold increase in polarisation” with growing divisions within countries due to the rise of opinion media, alongside a widening gap between open societies and regimes that control their media.

As regards misinformation, the Forum on Information and Democracy has developed 250 recommendations on how to stop “infodemics” (Box ‎4.4) and RSF also recommend several measures to enhance press freedom for OECD Members (Box ‎4.5).

Box ‎4.4. Recommendations from the Forum on Information and Democracy on countering mis- and disinformation and infodemics

False or manipulated information has proliferated steadily online, even before the COVID-19 pandemic, and has increased exponentially since, endangering democracies and human rights, including the right to health. Several countries, including France, Germany and Türkiye have already adopted regulations aimed at addressing this major challenge. Other jurisdictions, including Canada, the United Kingdom and the EU, have initiated discussion processes.

In a report entitled How to End Infodemics, the Forum on Information and Democracy – which was created by 11 organisations from civil society – made 250 recommendations centred on four major structural challenges: platform transparency, content moderation, promotion of reliable news and information, and private messaging services.

Platform transparency: Transparency obligations in public regulation on digital platforms that structure the information and communication space on their core functions should include content moderation, content ranking, content targeting and social influence building.
Content moderation: States should create a new model of meta-regulation with regard to content moderation while respecting a set of principles based on international human rights law: legality, necessity and proportionality, legitimacy, equality and non-discrimination. Nonetheless, they should refrain from establishing laws or arrangements that would require the “proactive” monitoring or filtering of content, which is both inconsistent with the right to privacy and likely to amount to prepublication censorship. They should also refrain from adopting models of regulation in which government agencies, rather than judicial authorities, become the arbiters of lawful expression.
Promotion of reliable news and information: At the core of any social platform is a technical construction resulting from deliberate choices in design, architecture and engineering. States could explore and develop a new regulatory focus on digital architecture and software engineering in the regulation of online service providers, with safety and quality standards developed in collaboration with experts. Online service providers should be subject to an obligation of neutrality in relation to their own interests and be required to represent reality honestly and not limit its representation to the content, goods or services they have an interest in. Platform conflicts of interest should therefore be prohibited so that the information and communication space is not governed or influenced by commercial, political or any other interests.
Private messaging services: Instead of just being used to exchange private messages, messaging apps are being used in some countries to massively disseminate disinformation. It would therefore seem essential to establish safeguards in private messaging services when they enter into a public space. New specific obligations should be imposed on service providers, in particular the obligation to create reporting mechanisms allowing users to report hateful or illegal content, in order to be able to take appropriate action, and the obligation to create effective mechanisms for appealing against moderation decisions.

Source: https://rsf.org/en/forum-information-and-democracy-250-recommendations-how-stop-infodemics

Box ‎4.5. RSF recommendations on enhancing press freedom for OECD Members

To foster an enabling environment for press freedom and reliable, fact-based journalism, RSF recommends introducing the following:

Mechanisms to protect journalists at risk, including:
- Protection programmes and the systematic investigation of threats or attacks against the press.
- At the international level, the establishment of a UN Special Representative of the Secretary-General to serve as a “protector of journalists”, ensure the implementation of existing UN mechanisms, hold states accountable and impose concrete costs on perpetrators.
A legal framework that creates an environment in which the press can perform its social role, including provisions that:
- Are in compliance with international standards, in particular Article 19 of the International Covenant on Civil and Political Rights and permitted restrictions on this freedom.
- Guarantee media pluralism and limiting concentrations in the media sector.
- Protect the independence of journalists and the media.
- Guarantee the right of journalists to protect the secrecy of their sources.
- Guarantee the independence of the regulatory authorities from political power.
- Guarantee transparency of public policies and the right to access public documents.
A legal framework that counters disinformation and ensures the future of journalism, including regulations that:
- Promote the transparency of platforms and the auditability of their algorithms.
- Secure the political, ideological and religious neutrality of platforms.
- Promote the discoverability of reliable information on the basis of self-regulatory standards defined by professional communities.
- Ensure that moderation of content respects international standards on freedom of expression.
- Safeguard media pluralism and competition by promoting a more pluralistic, open and decentralised digital environment.
- Guarantee the independence of national regulators and the future relevance of regulation.

4.4. Protection of online civic space

4.4.1. An open Internet as a facilitator of civic participation: A review of legal frameworks

An open Internet 9 has become a precondition for the enjoyment of civic space and the transformative nature of the Internet as an open platform, that facilitates citizen and stakeholder participation and dramatically expands civic freedoms, has been recognised by several international bodies. The UN Human Rights Council has recognised, for example, that “the same rights that people have offline must also be protected online, in particular freedom of expression” (UN, 2018[42]; UN, 2021[10]). OECD work in this area is under the purview of the OECD Committee on Digital Economy Policy (CDEP). In this regard, the OECD Recommendation of the Council on Principles for Internet Policy Making [OECD/LEGAL/0387], adopted in 2011, “recognises that the Internet provides an open, decentralised platform for communication, collaboration, innovation, creativity, productivity improvement and economic growth” (2011[43]). The Recommendation aims to preserve and promote the “open, distributed and interconnected nature of the Internet” (OECD, 2011[43]) and encourage governments to collaborate to safeguard personal data, protect intellectual property rights and ensure cybersecurity, while also maintaining respect for fundamental rights and the open nature of the Internet (OECD, 2014[44]). Infrastructure is key in terms of governments’ facilitation of access to an open Internet – and by extension, online civic space – and the OECD Recommendation on Broadband Connectivity [OECD/LEGAL/0322] recommends that adherents take measures to “eliminate digital divides and to reduce barriers to broadband deployment”, including by fostering the adoption and use of broadband services “at affordable prices, accessible for everyone, including all locations, genders, abilities, and socioeconomic circumstances” (OECD, 2021[45]).

The UN Human Rights Council has called upon countries to promote human rights-based, universal Internet access (UN, 2016[46]). The Council of Europe’s Committee of Ministers has similarly urged countries to create an enabling environment for Internet freedom (CoE, 2016[47]) and acknowledged their responsibility to take reasonable measures to protect and promote the universality, integrity and openness of the Internet as a means of safeguarding freedom of expression and access to information regardless of frontiers.10 EU Member states are also required to implement the Open Internet Regulation 2015/2120, according to which Internet service providers are prohibited from blocking or slowing down Internet traffic, except where necessary (EU Monitor, 2015[48]). Additionally, the OECD, the Council of Europe’s Committee of Ministers and other international bodies have emphasised that there should be no discrimination in the treatment of Internet traffic and data (net neutrality)11 (OECD, 2019[49]; UN et al., 2011[50]; CoE, 2016[51]). More recently in January 2022, the European Parliament, the European Council and the EC issued a European Declaration on Digital Rights and Principles for the Digital Decade, a non-binding guidance to a human-centric and rights-based approach to digital transformation (EC, 2022[52]).

The principle of an open Internet is established in law in most respondents (75% of all respondents, 94% of respondent OECD Members) (Figure ‎4.13). Some respondents do not explicitly protect the Internet in law but do have relevant legislation on freedom of expression and information. Costa Rica was one of the first to recognise Internet access as a fundamental right in 2010 (Freedom House, n.d.[53]). In Mexico, the right to access the Internet is set out in the constitution.

Figure ‎4.13. Legal provisions protecting the open Internet, 2020

Nevertheless, there are a number of common legally mandated exceptions and conditions related to an open Internet, often to protect other rights (Figure ‎4.14). For example, in 86% of all respondents and 86% of respondent OECD Members, the Internet can be restricted by law in the interests of individuals’ privacy rights. In 81% of all respondents and 86% of respondent OECD Members similar restrictions are possible to safeguard individuals’ intellectual property rights. The protection of children’s rights is an exception in a smaller group of countries (33% of all respondents, 25% of respondent OECD Members), where rules must be followed to ensure that publications or broadcasts do not include sexually offensive or abusive images of children. Furthermore, in 86% of all respondents and 93% of respondent OECD Members, laws or case law set out limitations in the interests of national sovereignty or security and 28% of all respondents (29% of OECD Members) have limitations to counter threats to public safety. Finally, in 14% of all respondents and 11% of OECD Members, the Internet can be limited to curb the spread of misinformation 12, and in 11% of all respondents (11% of OECD Members) restrictions are based on threat to public morals. Regarding the exceptions to this right, while some countries referred to specific laws that apply to online matters, other countries referred to general legislation, which applies both on and off line.

Figure ‎4.14. Legally mandated exceptions to the right to an open Internet, 2020

Several countries have a complaints system (either governmental or independent) for individuals regarding access to an open Internet, in addition to courts and NHRIs in place in all countries to address violations. For example, in Finland, the Finnish Transport and Communications Agency (Traficom) monitors net neutrality and ensures that the country adheres to relevant EU regulations. Citizens can contact Traficom if they have specific complaints about their operators which could not be solved between the two parties. The Bundesnetzagentur in Germany has similar responsibilities for monitoring and compliance, while the Ministry of Communication and Information has this role in Indonesia and can investigate any violations by service providers.

A number of OECD Members, among them Australia, Canada, Colombia, Israel and Korea, have passed legislation and taken other measures to facilitate and enhance access to and safety of the Internet and net neutrality, which are positive steps in terms of enhancing civic space online. At the same time, countries such as Australia have passed legislation permitting temporary or permanent restrictions on the Internet in the interests of combatting terrorist acts by obliging service providers to remove content considered to be harmful. A law in Germany likewise obliges online platforms to investigate and delete flagged content containing illegal content. The law has recently been amended to include the possibility of appeals proceedings in such cases (Library of Congress, 2021[54]). On the other hand, Latvia has adopted legislation that requires international service providers to block access to TV broadcasts that violate intellectual property rights (Dziadul, 2019[55]). In Colombia, a Supreme Court decision in December 2019 made blog and forum operators legally responsible for third-party defamatory user comments in the absence of measures to control them, according to Freedom House (2020[56]).

In very few OECD Members, such as Spain and Türkiye, legislation provides public authorities with wider powers to control Internet technologies and block access to Internet sites (EDRi, 2019[57]; Balamir Coskun, 2021[58]). Concerns have been voiced by civil society that in case of overly broad implementation, such laws can risk restricting access to the Internet and thus also civic space (Freedom House, 2021[59]). The UN Special Rapporteur on the promotion and protection of human rights and fundamental freedoms while countering terrorism has highlighted that electronic modes of expression are a critical means for civil society to exercise their freedom of expression and that restricting such platforms can affect civil society, journalists, human rights defenders and others disproportionally (Ní Aoláin, 2019[60]).

4.4.2. Implementation challenges and opportunities for freedom of expression online, as identified by CSOs and other stakeholders

Governments in OECD Members have generally protected and promoted freedom of expression online from interference such as Internet shutdowns. Internet shutdowns – the bluntest instrument involving a time or location-based large-scale Internet blackout – did not occur in 2019 in any of the 29 OECD Members that responded to a relevant survey question. Nevertheless, at the global level, Internet freedom declined for the 11^th consecutive year in 2021 according to Freedom House (2021[59]).

Government interference with overall access to the Internet, or to parts of the Internet through blocking or filtering access, can come in a wide range of forms. While most of Freedom House’s reports of government investigations, arrests or convictions of people for their social media posts between June 2020 and May 2021 concerned non-Members, such incidents were also reported in at least three OECD Members (Freedom House, 2021[61]; 2020[56]; 2021[62]; 2020[63]). Interference with access to an open Internet can also come in the form of network disruptions, such as slowdowns imposed on the Internet or parts of the Internet, or by blocking specific websites and applications or censoring specific online content. The V-Dem Institute’s indicator on government Internet filtering in practice,13 which is based on expert evaluation and measures how frequently governments censor political information on the Internet by blocking access to certain websites, shows that government censorship of the Internet and blocking access to certain websites never or almost never occurred in 2021 in the majority of respondent OECD Members (79%) that participated in the Survey. In 16% of OECD respondents, there have been a few occasions in which governments removed political content online. In one country, the government removed about half of the critical online political content, and in another, the government commonly removes online political content (V-Dem Institute, 2021[35]). According to a separate V-Dem Institute indicator on Internet shutdowns in practice, also based on expert assessment, 97% of OECD respondents never or almost never shut down the Internet in 2021 (2021[35]).

Freedom of expression online and the increased availability of digital tools and platforms to connect and share views remain particularly significant for youth and their ability to ensure their voices are heard. Box ‎4.6 highlights the importance of engaging this hyper-connected demographic, including in the context of the recovery from the COVID-19 crisis.

Box ‎4.6. Using digital tools to engage young people

The COVID-19 crisis has exacerbated pre-existing challenges for young people in terms of employment, education and mental health, while also curtailing civic space for youth 1 and youth-led organisations (OECD, 2020[64]). Promoting civic space is crucial to ensure that young people can meaningfully participate in public life and that public administrations are able to deliver on their diverse needs as they find themselves in a period of transition and stress. Broadening young people’s influence on decisions provides policy solutions based on a wider range of experiences and skills, it enhances young people’s trust in public institutions and supports policy outcomes that are responsive to all citizens (OECD, 2020[65]).

Media freedoms and protected online civic space are particularly relevant for young people, as they tend to use digital tools to inform and express themselves, communicate and associate more regularly than older age cohorts. Young people are more likely to use social media as their main source of news, thereby increasing their likelihood of being exposed to misinformation. According to a recent study, social media accounted for 88% of misinformation related to the COVID-19 pandemic between January and March 2020 (Brennen et al., 2020[66]). Young people are also more likely to use digital tools to voice their opinions and concerns: in 2018, 23% of people aged 15-29 surveyed across 22 OECD Members in the European Social Survey reported that they had shared or posted online about politics in the previous 12 months, compared to 15% of respondents aged 30 or more (OECD, 2020[65]). Furthermore, 75% of youth organisations surveyed 2 for the OECD report Governance for Youth, Trust and Intergenerational Justice: Fit for All Generations? believe online debates via social media will become even more important in the next five years (OECD, 2020[65]).

However, the COVID-19 crisis has also revealed vulnerabilities in terms of young people’s access to online tools. For instance, more than one in five 15-year-olds from socio-economically disadvantaged schools does not have access to a computer for schoolwork across OECD Members (OECD, 2020[67]). Policies and programmes to ensure more inclusive access to electronic devices, and connectivity among young users and to protect their civic space online are critical to overcoming the digital divide and countering misinformation. A number of countries are tackling these issues in the recovery from the COVID-19 crisis. For instance, a recent OECD analysis of recovery plans shows that numerous countries include, within their recovery plans, measures to promote digital literacy among young people (OECD, 2022[68]). In some cases, these efforts also aim to engage young and elderly people together to strengthen social cohesion: for instance, the Connected Lithuania programme was expanded to support projects led jointly by youth and senior organisations to promote digital literacy among marginalised communities (OECD, 2022[68]).

← 1. Mindful of the de-standardisation of life trajectories and the constant evolution and re-interpretation of particular stages of life, “youth” is defined as a period towards adulthood which is characterised by various transitions in one person’s life (e.g. from education to higher education and employment; from the parental home to renting an own apartment, etc.). Where possible, for statistical consistency, the UN classification of "youth" as individuals aged 15-24 is adopted.

← 2. Eighty-one youth organisations based in participating countries responded to the online survey that fed into the OECD report Governance for Youth, Trust and Intergenerational Justice: Fit for All Generations? (2020[65]). The survey was run between May 2019 and January 2020. Only the responses that included a valid URL/website presenting the work of a youth organisation were included in the final analysis (65 respondents).

Source: (Brennen et al., 2020[66]); (OECD, 2020[65]); (OECD, 2020[67]); (OECD, 2020[64]); (OECD, 2022[68])

Addressing online hate speech

Freedom of expression and pluralistic public opinion cannot be realised if individuals feel they must refrain from discussing certain topics or withdraw from public debate for fear of vilification or harmful racial, gender-based, or other stereotypes and discrimination (Illman, 2020[69]). In the digital era, concerns around hate speech have become increasingly pronounced. Hate speech is defined for the purposes of this report as any kind of communication in speech, writing or behaviour that attacks or uses pejorative or discriminatory language with reference to a person or a group on the basis of who they are, and aims to incite discrimination or violence towards that person or group, e.g. based on their religion, ethnicity, nationality, race, colour, descent, gender or other identity factors (Section 2.1 in Chapter 2 on legal frameworks governing hate speech whether on or off line). There is an emerging consensus among governments, technology companies and civil society on the need to consider new policy and regulatory frameworks that encourage the flow of factual information and protect individuals and society from the unchecked spread of hateful or illegal content, while also preserving users’ freedom of expression.

The actual extent of hate speech and content that promotes harassment remains uncertain as comprehensive and comparable data regarding complaints are lacking. Nevertheless, the use of electronic forms of communication, such as social media and technology platforms, has made different forms of harmful content more visible and easier to spread (UN, 2020[70]). Reported online hate cases include calls to violence, murder, rape and, in their most extreme forms, calls to commit atrocities (de Varennes, 2021[71]). Empirical analysis suggests that the overwhelming majority of hate speech on social media is targeted against minorities (de Varennes, 2021[71]) and that the COVID-19 pandemic has exacerbated this issue. For example, the UN has found that COVID-19 has given rise to a new wave of hate speech, discrimination and scapegoating of particular individuals and groups using derogatory, misogynistic, xenophobic, Islamophobic and anti-Semitic language and that online social media as well as mainstream media are being used to spread it (UN, 2020[70]).

The challenges around combatting online hate and potential policy responses have been widely discussed within OECD Members and by the CDEP, as well as by international and regional bodies in recent years, with the emergence of a number of measures to combat it (UNHCHR, 2013[72]; UN, 2019[73]; CoE, 2015[74]; ECtHR, 2020[75]; Lanza, 2019[76]; CoE, 2022[77]; Brookings Institution, 2021[78]; OECD, 2020[79]).14 At the regional level, the EU introduced a voluntary code of conduct on hate speech in 2016 (EU, 2016[80]). In 2022, the EU adopted the so-called Digital Services Act, which imposes legal obligations on social media platforms related to transparency and moderating illegal and harmful content (EC, 2020[81]). The European Court of Human Rights has also ruled that in cases involving hate speech or incitement to violence, declaring Internet news portals liable for failing to remove hate speech and generally unlawful content does not violate the respective companies’ right to freedom of expression.15 At the same time, the UN Special Rapporteur on the promotion and protection of the right to freedom of opinion and expression has emphasised that countries may only demand actions from Internet companies that are justified under and in compliance with international law. He recommends that countries only restrict online content based on a court order in accordance with relevant, fair trial standards and that they should refrain from imposing disproportionate sanctions on Internet intermediaries (Kaye, 2018[82]).

The OECD, through the CDEP, has addressed similar concerns in regard to the role and responsibilities of online platforms in terms of confirmation bias, content bubbles and societal polarisation. The 2019 OECD report An Introduction to Online Platforms and their Role in the Digital Transformation (2019[83]) suggests tackling the question of whether social media platforms should still be considered private spaces or whether they have actually become public spaces.

Some governments have undertaken measures, for example co-regulation between the government and the private sector, and self-regulation by the platforms themselves, alongside legal requirements to push platforms to filter unlawful content (OECD, 2019[83]). Co- and self-regulation can allow platforms themselves to identify emerging challenges and develop solutions in a quick and effective manner. However, governments have raised concerns that self-regulation by online platforms alone is often inadequate and not sufficient (OECD, 2019[83]).

The consequences of anti-terrorism legislation have been raised as another source of concern for free expression online. While the spread of terrorist and violent extremist content (TVEC) online has contributed to numerous attacks and intensified pressure on content-sharing services to do more to prevent or avoid TVEC on their platforms, observers such as the UN Office of the High Commissioner for Human Rights and CSOs have expressed concerns that anti-terrorism legislation in some countries negatively impacts people’s civic freedoms, including freedom of expression (RSF, 2021[84]; OHCHR, 2015[85]; OECD, 2021[86]). The CDEP is currently examining TVEC-related policies and procedures undertaken by online content-sharing services as part of a broader initiative to develop a Voluntary Transparency Reporting Framework, which was launched in 2022 (OECD, 2020[79]). This portal allows online platforms to submit their standardised transparency reports on TVEC policies and actions, aiming to build trust in online platforms by providing a baseline standard for transparency. The wider objective of the project is to increase the accountability of online platforms in both protecting human rights and ensuring that the Internet is a safe space for all (OECD, 2022[87]).

In recent years, governments have taken a variety of measures to tackle the phenomenon of online harm. Internet platforms have adopted content policies on hate speech, including banning users from posting or sharing unlawful or illegal speech. At the same time, such measures are also being contested for limiting free expression, especially when platforms use automated processes to identify hate speech (Kaye, 2019[88]). Indeed, a public consultation conducted by the EC in 2020 shows that citizens are demanding caution in using automated tools that risk removing legal content, potentially leading to unintentional and unjustified limitations on freedom of expression (EC, 2020[89]). In addition, they are increasingly of the view that regulatory oversight and auditing of platforms’ actions and risk assessments are crucial (EC, 2020[89]; Smith, 2018[90]). Identifying the scope of regulatory solutions and measures against hate speech, in addition to hateful speech that is harmful, albeit not illegal, remains a challenge for governments and platforms alike and is a matter of ongoing debate.

Government-led measures to combat online hate speech

In addition to legislation discussed in Section 2.1.1 in Chapter 2, a wide range of other government-led initiatives to combat hate speech has been introduced in OECD Members and beyond in recent years (Figure ‎4.15).

National strategies and action plans

The adoption of national strategies or action plans to combat hate speech or speech that promotes harassment, including online, can help countries to move from a fragmented approach to a more encompassing, co-ordinated, whole-of-government, long-term approach with the aim of ensuring that all individuals have the same opportunity to participate in public life as others:35% of OECD respondents and 21% of all respondents to the OECD Survey on Open Government have embedded measures to combat such online content within broader national strategies or action plans on cybersecurity or strategies to combat racism, extremism, xenophobia or radicalism. To varying degrees, such strategies envisage actions related to awareness-raising, research, training for officials, support to victims, self-regulation, introducing regulatory measures and strengthening the capacities of law enforcement structures.

Figure ‎4.15. Measures to counter online hate speech, 2020

Some such strategies include a situation analysis of trends regarding online hate speech. For example, the Czech Republic’s strategy to fight against hate speech assesses how harmful speech has transformed throughout the years from being a problem at the fringes of society to becoming a prevalent phenomenon of a more polarised society in general (Czech Ministry of Interior, 2020[91]). Others, including those in the Slovak Republic and Spain, stress the need for enhanced research (Spanish Ministry of Interior, 2019[92]; Slovak Ministry of the Interior, 2020[93]). Yet others, such as Germany’s Action Plan Against Racism, also point to the need for systematic education in schools to raise awareness (ECRI, 2020[94]).

Strategies also focus on the introduction of new legislation to regulate online hate speech. In this context, a handful of countries, including Finland (Box ‎4.7) and Ireland, have consulted their populations on the issue. In Ireland, the government conducted a wide public consultation in 2020, throughout which the necessary limits of future hate speech legislation regarding freedom of expression were discussed in detail. Throughout the consultations, a broad consensus emerged among participants and experts that new legislation on harmful speech needed to contain robust safeguards for freedom of expression, such as protections for reasonable and genuine contributions to literary, artistic, political, scientific or academic discourse, in addition to fair and accurate reporting (Department of Justice Ireland, 2020[95]).

Box ‎4.7. OECD’s Civic Space Scan of Finland: recommendations on measures to combat hate speech targeting people in public professions in Finland

In 2021, as part of the OECD Civic Space Scan of Finland (2021[96]), a deliberative Citizens’ Panel commissioned by the Ministries of Finance and Justice developed recommendations for measures that Finland should implement to protect people working in the public eye from hate speech while safeguarding a plurality of views and freedom of expression (Jäske et al., 2021[97]). Panel participants were recruited from a random sample of 3 000 people, representing as diverse a sample as possible of the population of Finland in terms of age, gender, education, language and geographical area.

The Citizens’ Panel proposed a total of 25 measures. These emphasise: awareness-raising on hate speech and online shaming; the importance of straightforward and clear definitions of relevant terms; the need for better government communication on the issue to the wider public; and the need for proportional penalties, prevention initiatives and sufficient resources; in addition to the responsibility of online platforms to play a role in countering the phenomenon. The greatest consensus and support from panellists, who voted on each of the recommendations, was on the following:

Increase effective dissemination of information on hate speech to citizens.
Include citizens’ voices in decision making through citizens’ panels at the municipal and state levels.
Develop guidelines for decision makers and those officials most susceptible to online shaming, to address situations where these become victims of hate speech and/or harassment.
Oblige employers to draw up clear instructions for possible cases of hate speech or harassment, both to intervene in the situation and support the victim.
Appoint responsible persons in organisations and give clear responsibilities in the work against hate speech and online shaming to individual persons in central government.
Increase resources and a centralised website on support services for victims.

The most controversial recommendations were those concerning penalties for hate crimes (with some participants favouring harsher penalties and others community service), user moderation and forming a counterforce to target and engage with Internet trolls.

Source: OECD (2021[96]), Civic Space Scan of Finland, https://doi.org/10.1787/f9e971bd-en; Jäske, M. et al. (2021[97]), Recommendations for measures to be taken in Finland to protect people in public professions from hate speech and to safeguard free expression of opinion, https://julkaisut.valtioneuvosto.fi/bitstream/handle/10024/162966/Citizens_Panel_on_the_Freedom_of_Expression-Final_Report.pdf.

Government-led task forces

Several governments have set up specific structures within their administrations to counter online hate speech and content that promotes harassment. Figure ‎4.15 shows that 35% of respondent OECD Members and 28% of all respondents have done so. For example, Australia established a eSafety Commissioner, which is a government agency mandated to keep its citizens safe online, with powers related to harmful online content (eSafety Commissioner, 2021[98]). Italy established a working group to develop policy recommendations to counter online hate speech. In 2020, the working group in Italy proposed drafting a strategy with measures centred on civic education, legal culture, research, information and communication, as well as regulatory reforms that clarify roles and responsibilities (Gruppo di lavoro sull’odio online, 2021[99]). In the Slovak Republic, the government set up a committee to prevent and eliminate racism, xenophobia, antisemitism and other forms of intolerance, which serves as a platform for co-ordinating activities and defining priorities. The committee has a dedicated working group on hate speech on the Internet (Slovak Ministry of Interior, 2021[100]).

Targeted public awareness campaigns

Fifty-four percent of respondent OECD Members and 37% of all respondents have initiated targeted public awareness campaigns (Figure ‎4.15). Public information and education campaigns are essential in combatting negative stereotypes of and discrimination against individuals on the basis of their protected characteristics. Such campaigns can equip individuals with greater confidence to identify and challenge manifestations of intolerance. In Canada, for example, the government funded a project entitled Block Hate: Building Resilience against Online Hate Speech, which examines hate speech trends across the country and works with experts to develop online tools and digital literacy training (YWCA Canada, 2020[101]). Initiatives launched by the eSafety Commissioner in Australia have provided stakeholders and the general public with tools for a deeper analysis of online safety issues, including harmful content, which provides guidance for students, parents, educators and community workers.

Targeted training for public officials

Targeted training for public officials and law enforcement bodies on online hate and content that promotes harassment has been rolled out in 31% of respondent OECD Members and 26% of all respondents (Figure ‎4.15). In some respondents, such as the Czech Republic, Latvia and Spain, this training was carried out jointly for law enforcement and civil society actors (ECRI, 2020[102]; 2019[103]; Spanish Ministry of Interior, 2019[92]). In Latvia, this training included non-governmental organisations directly linked to vulnerable groups. In the Slovak Republic, the national human rights institution, the Slovak National Centre for Human Rights, undertook educational activities for members of law enforcement bodies and the Ministry of Education (2019[104]). In Finland, training activities focus on improving reporting, enhancing capacities of the law enforcement agents and improving support structures for victims (Finish Ministry of Justice, 2019[105]). Hate crimes are part of the basic and continuous instruction for Finnish police and annual specialists’ training is offered by the Police University College. In 2021, the entire Finnish police force was required to take mandatory online training on equality, good relations and hate crime (OECD, 2021[96]).

Publication of disaggregated data

Comprehensive data collection systems, with fully disaggregated data by category of offence, type of hate motivation, target group as well as judicial follow-up and outcomes, are important tools to measure the true scale of the challenge, better understand emerging trends, improve the effectiveness of responses and raise awareness. While 65% of respondent OECD Members and 44% of all respondents track such information, data are not always sufficiently disaggregated, thereby hindering a detailed understanding of the issue that could guide interventions (Figure ‎4.15). In some OECD Members, such as Austria and Latvia, data on related complaints, crimes, investigations and prosecutions are recorded within national statistics on hate crimes (ECRI, 2020[106]; 2019[103]). When combined in this manner, it is not possible to differentiate the number of hate speech incidents as distinct from hate crimes, which is a broader category of crime covering, for example, physical violence and threats of violence. It also makes monitoring and tracking crimes more difficult for victims. Moreover, national statistics usually record only those forms of hate speech that are criminalised (ECRI, 2020[102]), precluding monitoring of the phenomenon in a more comprehensive manner.

Some countries publish an analysis of trends together with the number of online hate speech cases. In Denmark, for example, the annual national police report includes an analysis of hate speech trends, showing numbers disaggregated by motivation and type of bias, and comparing the data across years (National Police Denmark, 2019[107]). In Germany, apart from general statistics on hate speech, the bi-annual transparency reports that social network companies publish under the Network Enforcement Act (NetzDG) have proven to be an important source of data on the phenomenon (ECRI, 2020[94]).

Support for victims

A total of 65% of respondent OECD Members and 51% of all respondents have established hotlines or complaint mechanisms to report online hate speech and provide support for victims. While in some countries, there are dedicated reporting mechanisms 16 specifically for online hate speech – such as in Belgium or Germany – often incidents can be reported to existing police structures or support mechanisms for victims through independent bodies that handle human rights complaints. In Belgium, for example, the national equality body (Unia, 2021[108]) provides complaints channels for online hate speech. Other OECD Members have embedded reporting channels within the reporting mechanisms of the police, such as in Finland, Latvia, Portugal and Sweden.

Alongside government structures, in some OECD Members, such as Ireland and Latvia, CSOs also provide reporting channels for victims and other support. In Belgium, the NHRI supports victims by acting as a civil party in court proceedings. In Germany, a network of counselling services specifically for victims of online hate speech is provided through the initiative Zivile Helden (2021[109]). In Sweden, authorities are enhancing victim support structures for individuals who are exposed to threats and hatred in connection with their participation in public discourse (Government of Sweden, 2017[110]). Specialised hate crime units have also been created in Swedish police structures to assist victims, train fellow police officers and conduct outreach and confidence-building activities in local communities and among vulnerable groups, according to ECRI (2018[111]).

Where complaint mechanisms for hate speech exist, research suggests that under-reporting remains a significant challenge, often due to a lack of confidence in the police or justice system (de Varennes, 2021[71]; FRA, 2020[112]; 2013[113]; CoE, 2015[74]). Reasons for this are numerous, including: a lack of knowledge of complaint mechanisms; victims blaming themselves for the crime; fear of re-victimisation, such as retaliation by perpetrators; language barriers; or the fear of being deported in cases involving undocumented persons, among others (Bayer, 2020[114]). To tackle this challenge, the police in Denmark conducted outreach activities, including information meetings and dialogues with minority communities, to encourage more victims of hate crimes to report such incidents (ECRI, 2020[115]).

In recent years, CSOs in several countries have also started gathering statistics on hate speech. Given the challenge of under-reporting, co-operation between national authorities and CSOs – particularly those representing the interests of target groups – is a promising path to obtaining a realistic picture of the extent of the problem.

Regulations on content moderation and reporting requirements for social media platforms

The extent to which online platforms should be responsible for monitoring, removing or blocking user content that does not conform to legal requirements has been under increased focus in recent years (OECD, 2019[83]). In the context of the OECD Survey on Open Government, only Australia and Germany confirmed to have introduced regulations on content moderation for social media platforms (Figure ‎4.15), while the Slovak Republic strategy in the fight against radicalisation and extremism specifies that a new legislative obligation will be introduced for social network operators in order to systematically remove hate speech from their sites and to archive the removed evidence for investigators (Slovak Ministry of the Interior, 2020[93]; Government of Australia, 2021[116]). The law in Germany requires social media platforms to take down or block manifestly illegal fake news, hate speech and certain other unlawful content within 24 hours of receiving a complaint. Furthermore, a recent decision of the High Court of Australia confirmed that media companies were liable for defamatory comments by third parties posted in response to articles appearing on their Facebook pages (Karp, 2021[117]; High Court of Australia, 2021[118]).

Legislation adopted at the EU level in September 2022 to tackle the spread of illegal content online will be applicable in all EU Member states. The Digital Services Act sets obligations for social media platforms and obliges them to react quickly while respecting fundamental rights (EC, 2022[119]). However, some strict approaches have been criticised for their potential to limit freedom of expression given the short timeframes allotted to social media companies to identify this material (UN Human Rights Committee, 2021[120]). The OECD report An Introduction to Online Platforms and Their Role in the Digital Transformation notes the need for sufficient clarity and guidance from governments for the platforms responsible for carrying out filtering responsibilities so that they can comply with filtering requirements without pre-empting legitimate freedom of expression (OECD, 2019[83]).

Self-regulation mechanisms

Self-regulation and voluntary codes of conduct are additional, increasingly common approaches, with 15% of respondent OECD Members and 12% of all respondent countries reporting their use (Figure ‎4.15). Indeed, the OECD report An Introduction to Online Platforms and Their Role in the Digital Transformation suggests that private companies may often be better placed to understand problems that need regulatory attention and react to changes in their markets more quickly than public regulatory authorities (OECD, 2019[83]). Self-regulatory schemes can be adopted by private and public bodies – such as parliaments, political parties, business organisations, cultural and sports associations – by stressing that the use of hate speech by persons affiliated with them is unacceptable and taking action to prevent and sanction such use.

Box ‎4.8. Countering online abuse targeting women

As with other online hate, gendered online violence aims to silence, stigmatise and intimidate but focuses on women, girls and people who identify as female (Šimonović, 2018[121]; Article 19, 2020[122]) through misogynistic harassment, abuse and threats (Posetti, 2021[123]). Research from UNESCO suggests that 73% of women worldwide have experienced some form of online violence (UNESCO, 2015[124]). In EU Member states, one in ten women have reportedly experienced sexual harassment online (EC, 2019[125]; EIGE, 2017[126]; FRA, 2014[127]). In the United States, research indicates that while 21% of women aged 18 to 29 reported being sexually harassed online, this figure was 9% among men (Pew Research Center, 2017[128]). According to research from Australia, 76% of women under 30 have experienced online abuse (Hunt, 2016[129]).

Gendered online abuse against women activists and journalists has become more prominent in recent years, with a potentially grave impact on their ability to engage in public life.17 Data from UNESCO suggest that nearly three-quarters of women journalists globally have experienced online abuse (Posetti, 2021[123]). In addition, evidence suggests that the increase in online violence has resulted in women limiting their online participation, self-censoring or abandoning certain types of coverage altogether (Digital Rights Foundation, 2019[130]; Ferrier, 2018[131]; CPJ, 2019[132]; Fundacion Karisma, 2016[133]; Šimonović, 2018[121]).

Recent studies also show that such online violence easily moves offline, resulting in abuse and harassment of women and, in its most extreme forms, physical attacks (Posetti, 2021[123]) and even killings (Lawlor, 2020[134]). In 2017, the Committee to Protect Journalists (CPJ) reported that in at least 40% of cases, journalists who were murdered reported receiving threats, including online, before they were killed (Witchel, 2017[135]). Furthermore, according to the UN Office of the High Commissioner for Human Rights, women belonging to ethnic minorities and Indigenous women, lesbian, bisexual and transgender women, women with disabilities and women from marginalised groups are at even greater risk (Al Hussein, 2017[136]). A global survey by UNESCO of 714 self-selecting women journalists from 125 countries also showed that while 64% of white respondents had experienced online violence, the rates for minorities and marginalised women were significantly higher (81% identifying as Black, 86% identifying as Indigenous, and 88% identifying as Jewish) (Posetti, 2021[123]). The same survey also suggests that discrimination against women intersects with sexual orientation: while 72% of heterosexual women indicated they had been targeted in online attacks, the rates of exposure for those identifying as lesbian and bisexual women were 88% and 85% respectively (Posetti, 2021[123]).

As a response, at least 44% of respondent OECD Members and 51% of all respondents have introduced specific measures to address online hate speech that targets women (Figure ‎4.16). For example, Spain and Türkiye both train public officials to combat online violence against women and have targeted public awareness campaigns to assess harmful online content. Colombia, Costa Rica, Estonia, Israel, Italy, Portugal and Spain have introduced support structures for women victims of online violence through hotlines, complaint mechanisms or free legal advice. Columbia and Israel both publish data on online hate against women. Israel disaggregates the published data based on the categories of offence type, age, region and platforms where the online hate occurred.

Figure ‎4.16. Measures to counter online hate speech and/or content that promotes violence or harassment directed at women, 2020

Key measures to consider on countering hate speech or harassment online

- Assessing whether all groups or individuals, including women and minority communities, have the same opportunities to participate online without fear of abuse, harassment, threats or self-censorship and taking measures to counter any exclusion, including via targeted outreach efforts among affected communities.

- Considering adopting national strategies to counter online hate and harassment to implement a co-ordinated, whole-of-government and long-term approach to countering this growing phenomenon.

- Encouraging self-regulation by public and private entities by incentivising appropriate codes of internal conduct.

- Gathering and publishing disaggregated data on the phenomenon and actively supporting monitoring of related trends by civil society, equality bodies and national human rights institutions; supporting research related to how hateful content is created and spread, why and by whom, and which responses are most effective.

- Addressing under-reporting by raising awareness of relevant, accessible complaint mechanisms and remedies and co-operating in this respect with media and technology companies.

- Strengthening victim support systems including legal support, in consultation with specialist CSOs and targeted persons or groups.

Countering mis- and disinformation to support democracy and safe civic spaces online

The spread of mis- and disinformation 18 can distort democratic engagement and threaten the public interest information ecosystems that are crucial for healthy civic space and functioning democracies. By convincing people to believe incorrect messages that may demonise political opponents, reinforce polarisation or distort policy debates, as well as inhibit access to timely, relevant and accurate information and data, the amplification of mis- and disinformation can undermine the public’s willingness and ability to engage constructively in democratic debate.

Concerns about the risks posed by false information are at an all-time high. Indeed, 76% of respondents to the 2022 Edelman Trust Survey in 27 countries indicated that they worry about false information being used as a weapon (Edelman, 2022[137]). Such concerns illustrate the urgency and importance of ensuring governments have the capacity to respond to immediate threats posed by the spread of mis- and disinformation, as well as strengthen the wider public interest information ecosystems.

While mis- and disinformation are not new phenomena, the emergence of online communication spaces and social media platforms that allow for virtually anyone to instantaneously be a source of information (or misinformation) and amplify such content globally is a fundamental shift. The Internet has changed and facilitated the ability for content to be created and shared in ways that are only beginning to be understood (Lesher, Pawelec and Desai, 2022[138]). Changing business models in the media industry and increasing polarisation are additional systemic issues that both affect – and are affected by – how people get and share information and who and what they trust.

At the same time, these technological changes have provided new avenues to conduct public debate and can contribute to dynamic engagement. The growth in the diversity of sources and the opportunities to access global information provided by social media and the Internet also offer an essential counterweight to proscribed, anticompetitive or otherwise restricted media (particularly notable in the context of Russia’s aggression in Ukraine). Online communication technologies have also enabled governments, media and CSOs to more easily engage with citizens and for citizens to communicate with each other.

Nevertheless, because these same technologies can be used to threaten basic elements of democratic life, governments need to identify what constructive roles they can play in helping to build societal resilience to the challenges to civic space caused by mis- and disinformation. The breadth and depth of the mis- and disinformation challenge call for a wide range of measures driven by a whole-of-government and whole-of-society approach (OECD, forthcoming[139]). Current and proposed initiatives must reflect the multi-sectoral and systemic challenges faced and be developed and implemented in partnership with media and CSOs. To that end, understanding and building proactive roles for government responses, as well as identifying opportunities to develop and expand such a whole-of-society approach, will be crucial.

Maintaining freedom of expression and an open Internet means that mis- and disinformation will never disappear. A focus on responding to specific threats while reducing systematic risks to their spread, however, suggests a range of relevant responses. For example, the public communication function – which is distinct from political communication and is intended to deliver information, listen and respond to citizens in the service of the common good (OECD, 2021[140]) – will continue to be an important and strategic tool to strengthening public interest information ecosystems. Notably, recent OECD analysis suggests that there are opportunities for governments to use social media platforms more effectively and consciously to promote interactive communication online in ways that help counteract mis- and disinformation (OECD, 2021[140]).

Furthermore, the forthcoming OECD Principles of Good Practice for Public Communication Responses to Mis- and Disinformation explore in more detail how this function can facilitate rapid, strategic, inclusive and proactive responses to information challenges (forthcoming[141]). These principles examine the communication practices, institutional frameworks and interventions that foster an enabling ecosystem in which governments support a whole-of-society effort to support the flow of trustworthy information and data while mitigating the spread of false and misleading content.

Other efforts to build long-term resilience include media, information and digital literacy efforts, which can better equip citizens with the skills to differentiate between accurate and false or misleading information and increase awareness of their role in preventing its spread. Regarding regulatory responses, governments can adopt measures aimed at increasing online platforms’ transparency related to sources and content of disinformation campaigns, content take-downs and moderation activities, algorithmic design and impact, as well as beneficial ownership and disclosure of the entities sponsoring certain content (OECD, forthcoming[139]; Lesher, Pawelec and Desai, 2022[138]). Governments can also implement measures that are indirectly connected to mis- and disinformation but nevertheless have significant implications on the structural and economic drivers that affect its spread. Importantly, strengthening the press and news media sector through encouraging diversity, editorial independence and ensuring high-quality news provision, including through public service broadcasters, can help build the resilience of the media and information ecosystem more widely (OECD, forthcoming[139]). Additional initiatives to counteract mis- and disinformation include: developing content moderation policies in a multi-stakeholder process and with independent oversight; better integrating humans and technology in the fight against untruths online; and designing a measurement agenda to improve the evidence base and inform more targeted policies to stop the creators and spreaders of untruths (Lesher, Pawelec and Desai, 2022[138]).

While each country has a unique policy and information context, promoting freedom of speech and reinforcing the space for democratic debate and engagement are essential. Facilitating the independent role of civil society and media organisations will be critical. Given the speed of changes to the spaces in which people, organisations and institutions communicate, a whole-of-society effort will be required to counteract the threats posed and to reinforce public interest information ecosystems.

Key measures to consider on countering mis- and disinformation online

- Implementing policies to respond to threats as well as to build more resilient societies against mis- and disinformation by pursuing a whole-of-society approach to transparently and constructively counteract and prevent the spread of false and misleading information; improving media and information literacy; and supporting domestic and international collaboration to identify responses, among others.

- Building capacity for more proactive, responsive and effective public communication that provides factual information to populations, fills information voids and counteracts mis- and disinformation.

- Supporting the design of policy and regulatory measures to increase transparency and prevent the spread of false and misleading content.

- Identifying regulatory and policy responses that reduce economic and structural drivers of mis- and disinformation, for example via initiatives to promote more responsible behaviour of online platforms and identifying ways to strengthen the broader media and information ecosystem.

4.5. Personal data protection, artificial intelligence (AI) and civic space

Emerging technologies, big data analytics and AI enable governments and businesses alike to obtain fine-grained information about individuals. Indeed, for an increasing number of companies, personal data have become core to their business models, either in terms of selling such data to third parties, for advertising purposes or for tailoring their services. For governments, data collection is mostly carried out for the purpose of service provision (e.g. health, education or taxation) but data can also be held for criminal investigations or identification purposes. Data gathered and stored by governments can reveal a great deal of personal information about individuals, providing insights into private spheres of life, such as membership of organisations, participation in protests, religious or social affiliations, sexual orientation and health status and it is essential to protect such data from misuse as part of protecting civic space.

As with the protection of privacy more generally (Section 2.1.4 in Chapter 2), personal data protection supports an enabling environment in which citizens, journalists and civil society actors can gain access to information, express their views, operate freely and thrive without fear of arbitrary or unlawful intrusion or interference in their activities. The global trend towards greater data protection is reflected in regional instruments and guidelines on data protection, such as the Council of Europe’s Data Protection Convention (CoE, 2001[142]), the Inter-American Declaration of Principles on Freedom of Expression (OAS, 2000[143]), the African Union Convention on Cyber Security and Personal Data Protection (2014[144]) and the General Data Protection Regulation (GDPR) (EU, 2016[145]). Recent instruments such as the GDPR, the EU Directive on open data and the re-use of public sector information of 2019 and the European Data Governance Act (DGA) of 2022 are trailblazing data-related regulations worldwide (OECD, forthcoming[146]).

While data-driven technologies can be highly beneficial in terms of enhancing public services and identifying emerging societal needs (OECD, 2019[147]), the vast amounts of data collected as well as the potential linking of public datasets also come with risks (Al Hussein, 2018[148]), for example, data breaches resulting from accidents, the identification of individuals, malicious hacking and unauthorised access or disclosure, also increase (OECD, 2020[8]; Pillay, 2014[149]). Such breaches greatly diminish trust in the digital ecosystem and may affect how citizens and CSOs conduct their online activities, negatively impacting civic space. These trends have contributed to an evolving awareness of the need for greater personal data protection (OECD, 2021[150]). CSOs have particular concerns, with 7% of surveyed CSOs in the EU reporting a concern about surveillance by law enforcement (Section 5.6.4 in Chapter 5). Data governance and privacy have long been a focus of the CDEP and are also reflected in the establishment of a dedicated Working Party on Data Governance and Privacy in the Digital Economy. The 2017 and 2020 OECD Digital Economy Outlooks, prepared under the purview of the CDEP, find that individuals are increasingly concerned about the extensive use of their personal data by governments and private companies (OECD, 2017[151]; 2020[8]). Furthermore, the OECD Survey on Drivers of Trust in Public Institutions, undertaken under the purview of the Public Governance Committee (PGC), found that, on average, only half (51%) of the population in 22 surveyed countries trust their governments to use their personal data safely (OECD, 2022[152]). Thus, citizens are sharing more information about themselves than ever before but they are also increasingly seeking information on how it is handled, coupled with assurances that they maintain some control over how it is used (OECD, 2020[153]).

The OECD has played an important role in promoting data protection and privacy as fundamental values for governments as they embrace digitalisation (Box ‎4.9) on the basis that consideration must be given to protecting citizens’ rights, and especially their right to privacy, to prevent misuse (OECD, 2020[153]). The inherent tensions between personal data protection and data transparency, in addition to access to information (ATI) and freedom of expression, and the rights of the individual versus the common good present significant, unresolved challenges in the digitalised world and call for a balanced approach to data governance where the benefits of data access and sharing are maximised and the risks controlled (OECD, 2021[154]). For example, many countries take an approach to protecting personal data and ATI known as habeas data, meaning that the “right of individuals to access, update, rectify, and delete personal data collected by third parties and stored in databases” is considered a constitutional right (Data-Pop Alliance, 2022[155]).

Even with solid legal frameworks in place, the OECD has found that over-compliance with personal data protection regulations by public authorities can lead to restricted ATI following information requests (OECD, 2018[156]). Data protection laws can thus be used by authorities to prevent CSOs or individuals from pursuing public interest research or investigative reporting or to force journalists to reveal sources (Franz, Hayes and Hannah, 2020[157]). Increasingly, public bodies responsible for ATI are moving towards combining their role in ATI and on personal data protection. This is the case for instance, in Argentina, Belgium and Mexico. Although both topics are treated in separate legal frameworks in most countries and require different technical capacities and training, their proximity and complementarities are pushing governments towards centralising their role into a single institution. Dual responsibility can allow institutions to identify and exploit potential synergies between both policy areas to ensure that personal data and privacy are safeguarded while also allowing ATI (Box ‎4.9).

Box ‎4.9. OECD standards on data sharing, privacy and data ethics in the public sector

Governments are increasingly leveraging digital technologies to improve and streamline core government functions, to inform the design and delivery of policies and services and, where feasible and appropriate, to automate decision making using algorithms to process data at scale. In past years, societal demand for ethical frameworks to complement data protection and privacy regulations has increased.

The OECD Recommendation of the Council concerning Guidelines Governing the Protection of Privacy and Transborder Flows of Personal Data [OECD/LEGAL/0188], which was adopted in 1980 and amended in 2013, promotes the global free flow of information with guidance on how to protect individual civil liberties while collecting, processing and sharing personal data (2013[158]). In 2021, CDEP, though the Working Party on Data Governance, undertook a review of the implementation, dissemination and continued relevance of the Recommendation based in responses to a questionnaire, thematic expert roundtables, focused thematic reports, inputs form relevant work streams and discussions in conference calls, workshops and comment by an ad hoc informal advisory group of civil society, academic and private sector actors, noting the continued importance of the guidelines “as an international reference on minimum standards for privacy and personal data protection” (OECD, 2021[150]). The same year, the OECD Recommendation of the Council on Enhancing Access to and Sharing of Data [OECD/LEGAL/0463] was adopted (OECD, 2021[154]) to support governments in harnessing the potential of “personal, non-personal, open, proprietary, public and private” data while protecting the rights of individuals (OECD, 2021[159]). It also assists governments in fostering trust in the data ecosystem based on responsible data access and sharing while stimulating investment (OECD, 2021[154]).

In addition, and in response to challenges around the use of data and public trust, the OECD has elaborated ten Good Practice Principles for Data Ethics in the Public Sector to ensure that trust and human rights and values are at the core of digital government and data policies, strategies, projects and initiatives and that public integrity is upheld through specific actions (OECD, 2021[160]):

1. Manage data with integrity.
2. Be aware of and observe relevant government-wide arrangements for trustworthy data access, sharing and use.
3. Incorporate data ethical considerations into governmental, organisational and public sector decision-making processes.
4. Monitor and retain control over data inputs, in particular those used to inform the development and training of AI systems, and adopt a risk-based approach to the automation of decisions.
5. Be specific about the purpose of data use, especially in the case of personal data.
6. Define boundaries for data access, sharing and use.
7. Be clear, inclusive and open.
8. Publish open data and source code.
9. Broaden individuals’ and collectives’ control over their data.
10. Be accountable and proactive in managing risks.

Source: OECD (2021[160]), (OECD, 2013[158]), (OECD, 2021[154]).

4.5.1. The essential role of oversight in monitoring the protection of personal data

The protection of personal data can be effectively guaranteed through the existence of accessible institutional oversight mechanisms. Key international and regional instruments safeguarding the right to personal data protection, including the GDPR, the Council of Europe Convention for the Protection of Individuals with Regard to Automatic Processing of Personal Data, and the African Union Convention on Cyber Security and Personal Data Protection, require countries to establish independent supervisory authorities to monitor the implementation of their respective instruments. All OECD respondents have such supervisory bodies with total or partial independence from other state bodies. The vast majority of OECD Members responding to the OECD Survey on Open Government confirmed that these bodies handle complaints (as do central government authorities and domestic courts) and that they are mostly funded by general government revenues but with their own budget lines. All EU members are bound to have an independent data protection officer in place, based on the obligation to implement the GDPR.

4.5.2. AI in the public sector and civic space

AI 19 is also increasingly used in the private and public sectors to more effectively address pressing societal challenges. The OECD is closely monitoring AI developments across the globe through the OECD Artificial Intelligence Policy Observatory, with country-specific information on national AI strategy and policy initiatives. The OECD has developed a recommendation to promote innovative and trustworthy AI (Box ‎4.10) and discussions on the broader governance of AI and its use in the digital economy are conducted under the purview of the CDEP.

Box ‎4.10. OECD Recommendation of the Council on Artificial Intelligence

The Recommendation on Artificial Intelligence [OECD/LEGAL/0449] – the first intergovernmental standard on AI (OECD, 2019[161]) – was adopted by the OECD Ministerial Council in May 2019. The recommendation identifies five complementary values-based principles for the responsible use of trustworthy, human-centric AI, calling on relevant actors to promote and implement them. These principles are: inclusive growth, sustainable development and well-being; human-centred values and fairness; transparency and explainability; robustness, security and safety; and accountability.

Crucially, the Recommendation calls stakeholders to proactively steward AI that advances the inclusion of marginalised populations. As part of the principle of “human-centred values and fairness”, it makes reference to AI actors respecting “the rule of law, human rights and democratic values”, including “privacy and data protection, non-discrimination and equality”, all of which are essential preconditions for protected civic space (Chapter 2).

Source: OECD (2019[161]), “Recommendation of the Council on Artificial Intelligence”, OECD Legal Instruments, OECD/LEGAL/0449, OECD Publishing, Paris, https://legalinstruments.oecd.org/en/instruments/OECD-LEGAL-0449.

In the public sector, evidence shows that AI can adapt services to users’ needs and preferences, provide wider access to services such as education, safety and health and, overall, can increase citizen well-being (Ubaldi et al., 2019[162]). However, empirical evidence also highlights the potential for violations of rights by AI systems (Eubanks, 2018[163]; Richardson, Schultz and Crawford, 2019[164]) and concerns about how AI affects individual rights are growing, with implications for the protection of civic space (UN, 2020[165]; CoE, 2020[166]). For example, AI algorithms can transfer biases from the analogue to the digital world through the data they use (Mijatović, 2018[167]; Chander, 2020[168]). One of the key concerns surrounding their use in the public sector is the risk of discrimination, which can arise when machine-learning systems are fed data that only consider and reflect certain demographic groups or reproduce prejudices against such groups (FRA, 2019[169]). Algorithmic decision making can thus have a direct impact on the development of public policies, fair access to public services and can create barriers to the ability of all social demographics to participate in public life.

Threats to the right to non-discrimination have also been raised in relation to AI systems used in crime prevention and judicial proceedings. Law enforcement agencies increasingly use predictive policing through algorithmic processing of historical crime data and other sources to reveal patterns of criminal activity and identify targets for police intervention, for example (AlgorithmWatch, 2020[170]; Gonzalez Fuster, 2020[171]).20 The EU Agency for Fundamental Rights (FRA) has stressed that this practice risks being discriminatory, possibly reflecting biases by individual officers and data gaps due to chronic under-reporting of certain types of crimes (2020[172]). To effectively contest decisions based on the use of algorithms, it is essential for people to know when they are used, what information is underpinning relevant decisions, and how and where to complain in the event of a discriminatory outcome. A prominent concern about AI is the lack of transparency in the use of algorithmic decision making and the fact that in practice, decisions are difficult for citizens and stakeholders to challenge. Information about why certain data are collected or fed into algorithms often remains opaque, according to the FRA (2020[172]).21

Further concerns relate to ATI and freedom of expression, both of which can be affected by algorithms used by social media platforms and online search engines, hindering stakeholders’ ability to engage in diverse public debate. Research suggests that social media platforms can limit exposure to varied perspectives and facilitate the formation of groups of like-minded users (Cinelli et al., 2021[173]). Online search engines that use algorithms to index and rank content also means that users are less likely to reach content that is not highly ranked (Pasquale, 2016[174]), again limiting their ability to access information. This can have an impact on government outreach efforts – such as public consultations – and by extension on public policy making because it influences the quality of public debate and participation. Table ‎4.1 illustrates concretely how some Adherents are implementing the OECD’s recommendation on AI, in terms of respecting the rule of law, human rights and democratic values in the context of developing AI technologies.

Table ‎4.1. Common elements in national strategies on AI related to the protection of civic space, OECD Members, 2021
Element in the national AI strategy	Adherents	Count
Mentions potential risks to civic freedoms	Canada, Chile, Colombia, Czech Republic, Denmark, Finland, Germany, Ireland, Latvia, Netherlands, Norway, Poland, Portugal, Spain, Sweden, United Kingdom	16
Proposes development of ethics framework/commission	Canada, Colombia, Czech Republic, Denmark, Finland, Germany, Ireland, Lithuania, Korea, Netherlands, Norway, Poland, Portugal, Spain, Sweden, United Kingdom	16
Proposes concrete types of oversight and redresses mechanisms to protect civic freedoms	Canada, Czech Republic, Denmark, Latvia, Netherlands, Norway, Poland, Spain, Sweden, United Kingdom	10
Proposes public participation in the development and oversight of AI technologies	Colombia, Finland, Germany, Ireland, Lithuania, Portugal, Spain	7
Proposes training courses to make citizens more aware of AI and its risks and to improve inclusion in AI	Estonia, Finland, Germany, Ireland, Netherlands, Norway, Portugal	7
Engages in in-depth discussion on the impact of AI on civic freedoms	Chile, Denmark, Latvia, Netherlands, Spain, Sweden	6
Contains information on public participation (citizens or CSOs) in the development of the strategy	Canada, Chile, Czech Republic, Germany, Portugal	5
Total		19
Note: The table is based on in-depth research on primary sources (AI strategies), provided by the OECD Members and non-Members in the context of the 2020 OECD Survey on Open Government. The strategies included in the analysis are from: Canada, Chile, Colombia, the Czech Republic, Denmark, Estonia, Finland, Germany, Ireland, Korea, Latvia, Lithuania, the Netherlands, Norway, Poland, Portugal, Spain, Sweden and the United Kingdom (England and Wales). Data on Ireland is based on OECD desk research and was shared with it for validation.
Source: 2020 OECD Survey on Open Government.

A total of 81% of respondent OECD Members and 57% of all respondents to the 2020 Survey on Open Government had a national strategy for AI by the end of 2021. Common elements found in the 19 national strategies assessed are a focus on potential risks to civic freedoms and the intention to develop an ethics framework or commission to guide the development and application of AI, in particular in the public sector (84% respectively). Six of the strategies (Chile, Denmark, Latvia, the Netherlands, Spain and Sweden) include a deeper discussion on the impact of AI on civic freedoms. The rights most commonly discussed are the right to privacy and protection against discrimination, where discussions mainly focus on personal data protection, transparency and the explainability of algorithmic decision making. Public participation, either in the development of the strategy itself or in related activities, is mentioned in strategies in 7 OECD Members (37%). More than half of the strategies (53%) propose introducing concrete oversight and redress mechanisms with the aim of protecting civic freedoms.

Key measures to consider on protecting civic space in the context of personal data protection and AI

- Assessing the impact of legislation and policies governing personal data protection on ATI, privacy and civic freedoms, as part of creating and maintaining an enabling environment for civil society and citizens to participate in government policy and decision making.

- Establishing and adequately resourcing independent public institutions that address the misuse of personal data and automated decision-making complaints in the public sector and ensuring that the requisite legal structure and funding allow these institutions to be both independent and sustainable.

- Ensuring transparency regarding automated decision making in the public sector.

References

[144] African Union (2014), African Union Convention on Cyber Security and Personal Data Protection, https://au.int/en/treaties/african-union-convention-cyber-security-and-personal-data-protection (accessed on 25 May 2022).

[148] Al Hussein, Z. (2018), The right to privacy in the digital age, Report of the United Nations High Commissioner for Human Rights, https://undocs.org/A/HRC/39/29 (accessed on 13 June 2022).

[136] Al Hussein, Z. (2017), Report on the promotion, protection and enjoyment of human rights on the Internet: Ways to bridge the gender digital divide from a human rights perspective, Report of the United Nations High Commissioner for Human Rights, https://undocs.org/A/HRC/35/9.

[170] AlgorithmWatch (2020), Automating Society Report 2020, https://automatingsociety.algorithmwatch.org/wp-content/uploads/2020/12/Automating-Society-Report-2020.pdf.

[176] Allen, R. and D. Masters (2020), Meeting the new challenges to equality and non-discrimination from increased digitisation and the use of Artificial Intelligence, https://equineteurope.org/wp-content/uploads/2020/06/ai_report_digital.pdf.

[4] Anderson, M. (2018), How social media users have discussed sexual harassment since #MeToo went viral, https://www.pewresearch.org/fact-tank/2018/10/11/how-social-media-users-have-discussed-sexual-harassment-since-metoo-went-viral/ (accessed on 29 April 2021).

[122] Article 19 (2020), Online harassment and abuse against women journalists and major social media platforms, https://www.article19.org/wp-content/uploads/2020/11/Gender-Paper-Brief-2.pdf.

[58] Balamir Coskun, G. (2021), Turkey’s New Internet Law and Its Effects on Freedom of Media, Reset Dialogues, https://www.resetdoc.org/story/turkey-internet-law-freedom-media/.

[178] Barton, A. (2014), Violence and Harassment against Women in the News Media: A Global Picture, International Womens Media Foundation/International News Safety Institute, https://www.iwmf.org/wp-content/uploads/2018/06/Violence-and-Harassment-against-Women-in-the-News-Media.pdf.

[114] Bayer, J. (2020), Hate speech and hate crime in the EU and the evaluation of online content regulation approaches, European Parliament, https://www.europarl.europa.eu/RegData/etudes/STUD/2020/655135/IPOL_STU%282020%29655135_EN.pdf.

[66] Brennen, J. et al. (2020), Types, Sources, and Claims of COVID-19 Misinformation, Factsheet April 2020, https://reutersinstitute.politics.ox.ac.uk/sites/default/files/2020-04/Brennen%20-%20COVID%2019%20Misinformation%20FINAL%20(3).pdf.

[78] Brookings Institution (2021), History explains why global content moderation cannot work, https://www.brookings.edu/techstream/history-explains-why-global-content-moderation-cannot-work/ (accessed on 31 May 2022).

[168] Chander, S. (2020), Data racism: A new frontier, European Network Against Racism, https://www.enar-eu.org/Data-racism-a-new-frontier.

[173] Cinelli, M. et al. (2021), “The echo chamber effect on social media”, Proceedings of the National Academy of Sciences, Vol. 118/9, https://doi.org/10.1073/pnas.2023301118.

[77] CoE (2022), Recommendation CM/Rec(2022)16[1] of the Committee of Ministers to member States on combatting hate speech, Adopted by the Committee of Ministers on 20 May 2022, Council of Europe, https://search.coe.int/cm/Pages/result_details.aspx?ObjectId=0900001680a67955.

[166] CoE (2020), Recommendation on the human rights impacts of algorithmic systems, Council of Europe, https://www.coe.int/en/web/freedom-expression/committee-of-ministers-adopted-texts/-/asset_publisher/aDXmrol0vvsU/content/recommendation-cm-rec-2020-1-of-the-committee-of-ministers-to-member-states-on-the-human-rights-impacts-of-algorithmic-systems?_101_I.

[51] CoE (2016), Recommendation CM/Rec(2016)1 of the Committee of Ministers to member States on protecting and promoting the right to freedom of expression and the right to private life with regard to network neutrality, Council of Europe, https://rm.coe.int/CoERMPublicCommonSearchServices/DisplayDCTMContent?documentId=0900001680645b44.

[47] CoE (2016), Recommendation CM/Rec(2016)5[1] of the Committee of Ministers to member States in the field of media and information society, Council of Europe, https://rm.coe.int/CoERMPublicCommonSearchServices/DisplayDCTMContent?documentId=0900001680645b44.

[34] CoE (2016), Recommendation of the Committee of Ministers to member States on the protection of journalism and safety of journalists and other media actors, Council of Europe, https://www.coe.int/en/web/freedom-expression/committee-of-ministers-adopted-texts/-/asset_publisher/aDXmrol0vvsU/content/recommendation-cm-rec-2016-4-of-the-committee-of-ministers-to-member-states-on-the-protection-of-journalism-and-safety-of-journalists-and-other-media- (accessed on 15 December 2021).

[74] CoE (2015), European Commission against Racism and Intolerance: General Policy Recommendation No. 15 on Combating Hate Speech, Council of Europe, https://rm.coe.int/ecri-general-policy-recommendation-no-15-on-combating-hate-speech/16808b5b01.

[142] CoE (2001), “Additional Protocol to the Convention for the Protection of Individuals with regard to Automatic Processing of Personal Data regarding supervisory authorities and transborder data flows”, ETS No. 181, https://rm.coe.int/1680080626 (accessed on 25 May 2022).

[30] Committee to Protect Journalists (2022), Journalists killed, https://cpj.org/data/killed/?status=Killed&motiveConfirmed%5B%5D=Confirmed&motiveUnconfirmed%5B%5D=Unconfirmed&type%5B%5D=Journalist&type%5B%5D=Media%20Worker&start_year=1992&end_year=2022&group_by=year (accessed on 13 June 2022).

[7] Couldry, N. and U. Mejias (2019), “Data Colonialism: Rethinking Big Data’s Relation to the Contemporary Subject”, Television & New Media, Vol. 20/4, pp. 336–349, https://doi.org/10.1177/1527476418796632.

[23] Council of Europe (2022), Journalism in Situations of Conflict and Aggression, https://rm.coe.int/compilation-of-coe-standards-relating-to-journalism-in-situations-of-c/1680a5b775 (accessed on 13 June 2022).

[175] Council of Europe (2021), No Hate Speech Youth Campaign, https://www.coe.int/en/web/no-hate-campaign/reporting-to-national-bodies#{%2237118627%22:[7]}.

[27] CPJ (2021), 2021 Prison Census, Committee to Protect Journalists, https://cpj.org/reports/2021/12/number-of-journalists-behind-bars-reaches-global-high/ (accessed on 18 May 2022).

[29] CPJ (2021), Explore All CPJ Data, Committee to Protect Journalists, https://cpj.org/data/killed/?status=Killed&motiveConfirmed%5B%5D=Confirmed&motiveUnconfirmed%5B%5D=Unconfirmed&type%5B%5D=Journalist&start_year=1992&end_year=2022&group_by=year (accessed on 13 June 2022).

[11] CPJ (2021), New Greek criminal code amendment threatens journalists with jail for spreading ‘false’ information, https://cpj.org/2021/12/new-greek-criminal-code-amendment-threatens-journalists-with-jail-for-spreading-false-information/ (accessed on 29 November 2022).

[132] CPJ (2019), The threats follow us home: Survey details risks for women journalists in U.S., Canada, Committee to Protect Journalists, https://cpj.org/2019/09/canada-usa-female-journalist-safety-online-harassment-survey/ (accessed on 13 June 2022).

[28] CPJ (n.d.), Database, Committee to Protect Journalists, https://cpj.org/data/killed/?status=Killed&motiveConfirmed%5B%5D=Confirmed&type%5B%5D=Journalist&type%5B%5D=Media%20Worker&start_year=1992&end_year=2022&group_by=year (accessed on 29 November 2022).

[91] Czech Ministry of Interior (2020), Strategy on the fight against extremism and hate speech 2020-2026, https://www.mvcr.cz/clanek/extremismus-vyrocni-zpravy-o-extremismu-a-strategie-boje-proti-extremismu.aspx.

[155] Data-Pop Alliance (2022), Habeas Data and Personal Data Protection in Latin America, https://datapopalliance.org/habeas-data-and-personal-data-protection-in-latin-america/ (accessed on 1 September 2022).

[71] de Varennes, F. (2021), Report of the UN Special Rapporteur on minority issues, Office of the High Commissioner for Human Rights – United Nations, https://www.ohchr.org/EN/Issues/Minorities/SRMinorities/Pages/annual.aspx.

[95] Department of Justice Ireland (2020), Legislating for Hate Speech and Hate Crime in Ireland, http://www.justice.ie/en/JELR/Legislating_for_Hate_Speech_and_Hate_Crime_in_Ireland_Web.pdf/Files/Legislating_for_Hate_Speech_and_Hate_Crime_in_Ireland_Web.pdf.

[130] Digital Rights Foundation (2019), Female Journalists in New Media, Experiences, Challenges and Gender Approach, https://digitalrightsfoundation.pk/wp-content/uploads/2019/03/Female-journalists-in-new-media-experiences-challenges-and-a-gendered-approach.pdf.

[55] Dziadul, C. (2019), “Latvia clamps down on internet piracy”, Broadband TV News, https://www.broadbandtvnews.com/2019/01/10/latvia-clamps-down-on-internet-piracy/ (accessed on 13 June 2022).

[52] EC (2022), COM(2022) 28 final European Declaration on Digital Rights and Principles for the Digital Decade, European Commission, https://ec.europa.eu/commission/presscorner/detail/en/IP_22_452 (accessed on 13 June 2022).

[119] EC (2022), The Digital Services Act, European Commission, https://ec.europa.eu/info/strategy/priorities-2019-2024/europe-fit-digital-age/digital-services-act-ensuring-safe-and-accountable-online-environment_en (accessed on 13 June 2022).

[89] EC (2020), Open public consultation on the Digital Services Act, European Commission, https://ec.europa.eu/digital-single-market/en/news/summary-report-open-public-consultation-digital-services-act-package?cookies=disabled.

[81] EC (2020), Proposal for a regulation of the European Parliament and of the Council on a single market for digital services (Digital Services Act) and amending Directive 2000/31/EC, European Commission, https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52020PC0825&from=en.

[125] EC (2019), 2019 Report on equality between women and men in the EU, European Commission, https://op.europa.eu/en/publication-detail/-/publication/f3dd1274-7788-11e9-9f05-01aa75ed71a1/?cookies=disabled.

[115] ECRI (2020), ECRI Conclusions on the implementation of recommendations in respect of Denmark, European Commission against Racism and Intolerance, https://rm.coe.int/ecri-conclusions-on-the-implementation-of-the-recommendations-in-respe/16809e8271.

[106] ECRI (2020), ECRI Report on Austria (sixth monitoring cycle), European Commission against Racism and Intolerance, https://rm.coe.int/report-on-austria-6th-monitoring-cycle-/16809e826f.

[94] ECRI (2020), ECRI Report on Germany, European Commission against Racism and Intolerance, https://rm.coe.int/ecri-report-on-germany-sixth-monitoring-cycle-/16809ce4be.

[102] ECRI (2020), ECRI Report on the Czech Republic, European Commission against Racism and Intolerance, https://rm.coe.int/ecri-6th-report-on-the-czech-republic/1680a0a086.

[103] ECRI (2019), ECRI report on Latvia (fifth monitoring cycle), European Commission against Racism and Intolerance, https://rm.coe.int/fifth-report-on-latvia/1680934a9f.

[111] ECRI (2018), ECRI Report on Sweden, European Commission against Racism and Intolerance, https://rm.coe.int/fifth-report-on-sweden/16808b5c58.

[75] ECtHR (2020), Guide on Article 10 of the European Convention on Human Rights, European Court of Human Rights, https://www.echr.coe.int/Documents/Guide_Art_10_ENG.pdf.

[137] Edelman (2022), Edelman Trust Barometer 2022, https://www.edelman.com/sites/g/files/aatuss191/files/2022-01/2022%20Edelman%20Trust%20Barometer%20FINAL_Jan25.pdf (accessed on 25 May 2022).

[57] EDRi (2019), Spain: New law threatens Internet freedoms, European Digital Rights, https://edri.org/our-work/spain-new-law-threatens-internet-freedoms/.

[126] EIGE (2017), Cyber violence against women and girls, European Institute for Gender Equality, https://eige.europa.eu/publications/cyber-violence-against-women-and-girls.

[98] eSafety Commissioner (2021), eSafety Commissioner, https://www.esafety.gov.au/key-issues/Illegal-harmful-content.

[80] EU (2016), Code of Conduct on Countering Illegal Hate Speech Online, European Union, https://ec.europa.eu/newsroom/just/document.cfm?doc_id=42985.

[145] EU (2016), Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, European Union, https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A02016R0679-20160504 (accessed on 25 May 2022).

[48] EU Monitor (2015), Regulation 2015/2120: Measures concerning open internet access, https://www.eumonitor.eu/9353000/1/j9vvik7m1c3gyxp/vjzq8r2gblvf (accessed on 25 May 2022).

[163] Eubanks, V. (2018), Automating Inequality: How High-Tech Tools Profile, Police, and Punish the Poor, Macmillan, New York, https://us.macmillan.com/books/9781250074317.

[33] European Commission (2021), Recommendation on the protection, safety and empowerment of journalists, https://digital-strategy.ec.europa.eu/en/library/recommendation-protection-safety-and-empowerment-journalists (accessed on 28 November 2022).

[105] Finish Ministry of Justice (2019), Information and case examples hate speech and hate crimes, https://yhdenvertaisuus.fi/documents/5232670/13949561/K%C3%A4sikirja+syytt%C3%A4jille+ja+tuomareille/e03c873a-e85d-a95e-e4cc-f677a55f4340/K%C3%A4sikirja+syytt%C3%A4jille+ja+tuomareille.pdf/K%C3%A4sikirja+syytt%C3%A4jille+ja+tuomareille.pdf.

[1] Forum on Information and Democracy (2021), Working Group on the Sustainability of Journalism: A New Deal for Journalism, Forum on Information and Democracy, https://informationdemocracy.org/wp-content/uploads/2021/06/ForumID_New-Deal-for-Journalism_16Jun21.pdf.

[172] FRA (2020), Getting the future right – Artificial intelligence and fundamental rights, European Union Agency for Fundamental Rights, https://fra.europa.eu/en/publication/2020/artificial-intelligence-and-fundamental-rights.

[112] FRA (2020), Roma and travellers in six countries, European Union Agency for Fundamental Rights, https://fra.europa.eu/en/publication/2020/roma-travellers-survey.

[169] FRA (2019), Data quality and artificial intelligence – mitigating bias and error to protect fundamental rights, European Union Agency for Fundamental Rights, https://fra.europa.eu/sites/default/files/fra_uploads/fra-2019-data-quality-and-ai_en.pdf.

[127] FRA (2014), Violence against women: An EU-wide survey – Main results, European Union Agency for Fundamental Rights, https://fra.europa.eu/en/publication/2014/violence-against-women-eu-wide-survey-main-results-report.

[113] FRA (2013), European Union lesbian, gay, bisexual and transgender survey: Results at a glance, European Union Agency for Fundamental Rights, https://fra.europa.eu/sites/default/files/eu-lgbt-survey-results-at-a-glance_en.pdf.

[157] Franz, V., B. Hayes and L. Hannah (2020), Civil Society Organizations and General Data Protection Regulation compliance: Challenges, Opportunities, and Best Practices, Open Society Foundations, London, https://www.opensocietyfoundations.org/publications/civil-society-organizations-and-general-data-protection-regulation-compliance#publications_download.

[62] Freedom House (2021), Freedom on the Net 2021: Mexico, https://freedomhouse.org/country/mexico/freedom-world/2021.

[61] Freedom House (2021), Freedom on the Net 2021: Turkey, https://freedomhouse.org/country/turkey/freedom-net/2021.

[59] Freedom House (2021), Freedom on the Net. The Global Drive to Control Big Tech, https://freedomhouse.org/sites/default/files/2021-09/FOTN_2021_Complete_Booklet_09162021_FINAL_UPDATED.pdf.

[56] Freedom House (2020), Freedom on the Net 2020: Colombia, https://freedomhouse.org/country/colombia/freedom-net/2020#footnote2_czsktln.

[63] Freedom House (2020), Freedom on the Net 2020: The Pandemic’s Digital Shadow, https://freedomhouse.org/report/freedom-net/2020/pandemics-digital-shadow.

[53] Freedom House (n.d.), Costa Rica: Freedom on the Net 2021 Country Report, https://freedomhouse.org/country/costa-rica/freedom-net/2021 (accessed on 25 May 2022).

[3] Freelon, D. (2016), Beyond the Hashtags: #Ferguson, #BlackLivesMatter, and the Struggle for Offline Justice, Center for Media and Social Impact at the School of Communication at American University, http://archive.cmsimpact.org/sites/default/files/beyond_the_hashtags_2016.pdf.

[133] Fundacion Karisma (2016), Misogyny in the Internet, Colombia, https://web.karisma.org.co/misoginia-en-internet-bombardeo-a-campo-abierto-contra-las-periodistas/.

[171] Gonzalez Fuster, G. (2020), Artificial Intelligence and Law Enforcement: Impact on Fundamental Rights, European Parliament, https://www.europarl.europa.eu/RegData/etudes/STUD/2020/656295/IPOL_STU(2020)656295_EN.pdf.

[116] Government of Australia (2021), Online Safety Act 2021, https://www.legislation.gov.au/Details/C2021A00076 (accessed on 31 May 2022).

[12] Government of Colombia (2000), Decree 1592 of 2000, https://www.suin-juriscol.gov.co/viewDocument.asp?id=1314526 (accessed on 15 December 2021).

[110] Government of Sweden (2017), Defending Free Speech - Measures to protect journalists, elected representatives and artists from exposure to threats and hatred, https://www.government.se/4990f9/contentassets/bd181f7b0f4640e7920807d110b3c001/action-plan-defending-free-speech.pdf.

[99] Gruppo di lavoro sull’odio online (2021), Odio online - Rapporto finale (Online Hate - Final Report), https://assets.innovazione.gov.it/1613027971-odio-report-6.pdf.

[118] High Court of Australia (2021), Fairfax Media Publications Pty Ltd v Voller, Nationwide News Pty Limited v Voller and Australian News Channel Pty Ltd v Voller, Judgment of 8 September 2021, http://hcourt.gov.au.

[129] Hunt, E. (2016), “Online harassment of women at risk of becoming ’established norm’, study finds”, The Guardian, https://www.theguardian.com/lifeandstyle/2016/mar/08/online-harassment-of-women-at-risk-of-becoming-established-norm-study.

[19] ICNL (2021), Legal Responses to Disinformation, International Center for Not-for-Profit Law, https://www.icnl.org/wp-content/uploads/2021.03-Disinformation-Policy-Prospectus-final.pdf.

[180] IFJ (2021), International Federation of Journalists, International Federation of Journalists, https://www.ifj.org/media-centre/news/detail/category/press-releases/article/45-journalists-and-media-staff-killed-in-work-related-incidents-during-2021-says-ifj.html#:~:text=The%20International%20Federation%20of%20Journalists,recorded%20last%20year%20in%.

[69] Illman, M. (2020), Järjestelmällinen häirintä ja maalittaminen: Lainsäädännön arviointia [Systematic harassment and targeting: Legislative assessment (unofficial title translation)], Government report 2020:3, Government of Finland, https://julkaisut.valtioneuvosto.fi/bitstream/handle/10024/162579/3-2020-VN%20Selvitys.pdf.

[97] Jäske, M. et al. (2021), Recommendations for measures to be taken in Finland to protect people in public professions from hate speech and to safeguard free expression of opinion, Ministry of Finance, Ministry of Justice, Åbo Akademi University, Finland, https://julkaisut.valtioneuvosto.fi/bitstream/handle/10024/162966/Citizens_Panel_on_the_Freedom_of_Expression-Final_Report.pdf.

[117] Karp, P. (2021), “High court rules Australian media companies can be liable for defamatory comments posted on Facebook pages”, The Guardian, https://www.theguardian.com/media/2021/sep/08/high-court-rules-australian-media-companies-are-liable-for-defamatory-comments-posted-on-their-facebook-pages.

[88] Kaye, D. (2019), Promotion and protection of the right to freedom of opinion, United Nations, Geneva, https://documents-dds-ny.un.org/doc/UNDOC/GEN/N19/308/13/PDF/N1930813.pdf?OpenElement.

[82] Kaye, D. (2018), Report of the Special Rapporteur on the promotion and protection of the right to freedom of opinion and expression, A/HRC/38/35, United Nations, https://www.undocs.org/A/HRC/38/35.

[76] Lanza, E. (2019), Informe Anual de la Relatoria Especial para la libertad de Expresion, Comisión Interamericana de Derechos Humanos, http://www.oas.org/es/cidh/expresion/informes/ESPIA2019.pdf.

[134] Lawlor, M. (2020), Final warning: death threats and killings of human rights defenders, Special Rapporteur on the situation of human rights defenders, Office of the High Commissioner for Human Rights, https://www.ohchr.org/EN/Issues/SRHRDefenders/Pages/AnnualReports.aspx.

[138] Lesher, M., H. Pawelec and A. Desai (2022), Disentangling untruths online: Creators, spreaders and how to stop them, OECD Publishing, Paris, https://doi.org/10.1787/84b62df1-en (accessed on 25 May 2022).

[54] Library of Congress (2021), Germany: Network Enforcement Act Amended to Better Fight Online Hate Speech, https://www.loc.gov/item/global-legal-monitor/2021-07-06/germany-network-enforcement-act-amended-to-better-fight-online-hate-speech/.

[2] Matasick, C., C. Alfonsi and A. Bellantoni (2020), “Governance responses to disinformation : How open government principles can inform policy options”, OECD Working Papers on Public Governance, No. 39, OECD Publishing, Paris, https://doi.org/10.1787/d6237c85-en.

[167] Mijatović, D. (2018), Safeguarding human rights in the era of artificial intelligence, Council of Europe, https://www.coe.int/en/web/commissioner/-/safeguarding-human-rights-in-the-era-of-artificial-intelligence.

[15] Ministry of Human Rights (2018), Ordinance Nº 300 of 3 September 2018, Ministry of Human Rights, https://www.in.gov.br/materia/-/asset_publisher/Kujrw0TZC2Mb/content/id/39528373/do1-2018-09-04-portaria-n-300-de-3-de-setembro-de-2018-39528265.

[131] Munoz, E. (ed.) (2018), Attacks and Harassment: The Impact on Female Journalists and Their Reporting, International Women’s media Foundation, https://www.iwmf.org/wp-content/uploads/2018/09/Attacks-and-Harassment.pdf.

[107] National Police Denmark (2019), Hate crimes in 2019: National Police’s annual report on hate crimes, https://politi.dk/-/media/mediefiler/landsdaekkende-dokumenter/statistikker/hadforbrydelser/hadforbrydelser-2019.pdf?la=da&hash=86E26E960D6896CD33EBBB8ECE8487499BF986D3.

[60] Ní Aoláin, F. (2019), Impact of measures to address terrorism and violent extremism on civic space and the rights of civil society actors and human rights defenders, United Nations, https://undocs.org/Home/Mobile?FinalSymbol=A%2FHRC%2F40%2F52&Language=E&DeviceType=Desktop&LangRequested=False.

[16] Norwegian Ministry of Culture and Equality (2020), The Media Liability Act, https://lovdata.no/dokument/NLE/lov/2020-05-29-59.

[26] OAS (2020), Joint declaration: Governments must promote and protect access to and free flow of information during pandemic, Organization of American States, https://www.oas.org/en/iachr/expression/showarticle.asp?artID=1170&lID=1 (accessed on 15 December 2021).

[143] OAS (2000), Declaration of Principles on Freedom of Expression, Organization of American States, Washington DC, https://www.oas.org/en/iachr/expression/showarticle.asp?artID=26.

[152] OECD (2022), Building Trust to Reinforce Democracy: Main Findings from the 2021 OECD Survey on Drivers of Trust in Public Institutions, Building Trust in Public Institutions, OECD Publishing, Paris, https://doi.org/10.1787/b407f99c-en.

[68] OECD (2022), “Delivering for youth: How governments can put young people at the centre of the recovery”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/92c9d060-en.

[87] OECD (2022), Voluntary Transparency Reporting Framework, OECD, Paris, https://www.oecd.org/digital/vtrf/ (accessed on 6 May 2022).

[96] OECD (2021), Civic Space Scan of Finland, OECD Public Governance Reviews, OECD Publishing, Paris, https://doi.org/10.1787/f9e971bd-en.

[182] OECD (2021), “DAC Recommendation on Enabling Civil Society in Development Co-operation and Humanitarian Assistance”, OECD Legal Instruments, OECD/LEGAL/5021, OECD, Paris, https://legalinstruments.oecd.org/en/instruments/OECD-LEGAL-5021.

[159] OECD (2021), Data governance: Enhancing access to and sharing of data, OECD, Paris, https://www.oecd.org/sti/ieconomy/enhanced-data-access.htm (accessed on 13 May 2022).

[160] OECD (2021), Good Practice Principles for Data Ethics in the Public Sector, OECD, Paris, https://www.oecd.org/digital/digital-government/good-practice-principles-for-data-ethics-in-the-public-sector.htm.

[140] OECD (2021), OECD Report on Public Communication: The Global Context and the Way Forward, OECD Publishing, Paris, https://doi.org/10.1787/22f8031c-en (accessed on 25 May 2022).

[45] OECD (2021), “Recommendation of the Council on Broadband Activity”, OECD Legal Instruments, OECD-LEGAL-0322, OECD, Paris, https://legalinstruments.oecd.org/en/instruments/OECD-LEGAL-0322.

[154] OECD (2021), “Recommendation of the Council on Enhancing Access to and Sharing of Data”, OECD Legal Instruments, OECD/LEGAL/0463, OECD, Paris, https://legalinstruments.oecd.org/en/instruments/OECD-LEGAL-0463.

[150] OECD (2021), Report on the implementation of the Recommendation of the Council concerning guidelines governing the protection of privacy and transborder flows of personal data, C(2021)42, OCDE, Paris, https://one.oecd.org/document/C(2021)42/en/pdf (accessed on 13 May 2022).

[86] OECD (2021), Webinar briefing: The impact of national and global security measures on civic space, OECD, Paris, https://www.oecd.org/gov/open-government/impact-national-and-global-security-measures-on-civic-space-summary-report.pdf.

[79] OECD (2020), “Current approaches to terrorist and violent extremist content among the global top 50 online content-sharing services”, OECD Digital Economy Papers, No. 296, OECD Publishing, Paris, https://doi.org/10.1787/68058b95-en.

[5] OECD (2020), “Digital transformation and the futures of civic space to 2030”, OECD Development Policy Papers, No. 29, OECD Publishing, Paris, https://doi.org/10.1787/79b34d37-en.

[153] OECD (2020), “Going Digital Integrated Policy Framework”, OECD, Paris, http://www.oecd.org/going-digital (accessed on 16 May 2022).

[65] OECD (2020), Governance for Youth, Trust and Intergenerational Justice: Fit for All Generations?, OECD Public Governance Reviews, OECD Publishing, Paris, https://doi.org/10.1787/c3e5cb8a-en.

[67] OECD (2020), “Learning remotely when schools close: How well are students and schools prepared? Insights from PISA”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/3bfda1f7-en.

[8] OECD (2020), OECD Digital Economy Outlook 2020, OECD Publishing, Paris, https://doi.org/10.1787/bb167041-en.

[37] OECD (2020), OECD Public Integrity Handbook, OECD Publishing, Paris, https://doi.org/10.1787/ac8ed8e8-en.

[64] OECD (2020), “Youth and COVID-19: Response, recovery and resilience”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/c40e61c6-en.

[83] OECD (2019), An Introduction to Online Platforms and Their Role in the Digital Transformation, OECD Publishing, Paris, https://doi.org/10.1787/53e5f593-en.

[147] OECD (2019), Enhancing Access to and Sharing of Data: Reconciling Risks and Benefits for Data Re-use across Societies, OECD Publishing, Paris, https://doi.org/10.1787/276aaca8-en.

[49] OECD (2019), Going Digital: Shaping Policies, Improving Lives, OECD Publishing, Paris, https://doi.org/10.1787/9789264312012-en.

[161] OECD (2019), “Recommendation of the Council on Artificial Intelligence”, OECD Legal Instruments, OECD/LEGAL/0449, OECD Publishing, Paris, https://legalinstruments.oecd.org/en/instruments/OECD-LEGAL-0449.

[156] OECD (2018), Open Government Data Report: Enhancing Policy Maturity for Sustainable Impact, OECD Digital Government Studies, OECD Publishing, Paris, https://doi.org/10.1787/9789264305847-en.

[151] OECD (2017), OECD Digital Economy Outlook 2017, OECD Publishing, Paris, https://doi.org/10.1787/9789264276284-en.

[44] OECD (2014), OECD Principles for Internet Policy Making, OECD, Paris, https://www.oecd.org/digital/ieconomy/oecd-principles-for-internet-policy-making.pdf.

[158] OECD (2013), “Recommendation of the Council concerning Guidelines Governing the Protection of Privacy and Transborder Flows of Personal Data”, OECD Legal Instruments, OECD-LEGAL-0188, OECD, Paris, https://legalinstruments.oecd.org/en/instruments/OECD-LEGAL-0188.

[43] OECD (2011), “OECD Recommendation of the Council on Principles for Internet Policy Making”, OECD Legal Instruments, OECD/LEGAL/0387, OECD, Paris, https://legalinstruments.oecd.org/en/instruments/OECD-LEGAL-0387.

[139] OECD (forthcoming), Background note on Public Governance for Combating Mis-information and Dis-information for the Meeting of the Public Governance Committee at Ministerial level on “Building Trust and Reinforcing Democracy”, GOV/PGC(2022)8.

[146] OECD (forthcoming), Going Digital in Latin America.

[141] OECD (forthcoming), “Good Practice Principles for Public Communication Responses to Mis- and Disinformation”, https://www.oecd.org/governance/reinforcing-democracy/ (accessed on 25 May 2022).

[85] OHCHR (2015), Two legal reform projects undermine the rights of assembly and expression in Spain - UN experts, Office of the High Commissioner for Human Rights – United Nations, https://www.ohchr.org/EN/NewsEvents/Pages/DisplayNews.aspx?NewsID=15597.

[18] Parliament of Australia (n.d.), Criminal Code, https://www.aph.gov.au/Parliamentary_Business/Committees/Joint/Intelligence_and_Security/ISEastAsia/Report/Section?id=committees%2Freportjnt%2F024152%2F25821 (accessed on 29 November 2022).

[174] Pasquale, F. (2016), “Platform Neutrality: Enhancing Freedom of Expression in Spheres of Private Power”, U of Maryland Legal Studies Research Paper 2016-24, https://ssrn.com/abstract=2779270.

[128] Pew Research Center (2017), Online Harassment 2017, http://assets.pewresearch.org/wp-content/uploads/sites/14/2017/07/10151519/PI_2017.07.11_Online-Harassment_FINAL.pdf.

[149] Pillay, N. (2014), Report of the Office of the United Nations High Commissioner for Human Rights. The right to privacy in the digital age, United Nations, https://undocs.org/A/HRC/27/37.

[123] Posetti, J. (2021), The Chilling: Global trends in online violence against women journalists, UNESCO, UNESCO, Paris, https://www.icfj.org/sites/default/files/2021-04/The%20Chilling_POSETTI%20ET%20AL_FINAL.pdf.

[177] Reporters Without Borders (2021), German BND Act: A missed opportunity for press freedom, https://rsf.org/en/news/german-bnd-act-missed-opportunity-press-freedom.

[17] Reporters Without Borders (2021b), Norway, https://rsf.org/en/norway.

[39] Reporters without Borders (2021), 2021 World Press Freedom Index, https://rsf.org/en/ranking/2021.

[32] RFI (2021), Violent protests, harassment harm press freedom in France, https://www.rfi.fr/en/france/20210420-violent-protests-harassment-harm-press-freedom-in-france-watchdog-says-journalism-reporters-without-borders-police-media-ownership (accessed on 28 February 2022).

[164] Richardson, R., J. Schultz and K. Crawford (2019), “Dirty Data, Bad Predictions: How Civil Rights Violations Impact Police Data, Predictive Policing Systems, and Justice”, 94 N.Y.U. L. REV. ONLINE 192, https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3333423.

[38] RSF (2021), Detailed methodology, Reporters Without Borders, https://rsf.org/en/detailed-methodology (accessed on 29 November 2021).

[84] RSF (2021), “Journalists threatened with imprisonment under Turkey’s terrorism law”, Reporters Without Borders, https://rsf.org/en/news/journalists-threatened-imprisonment-under-turkeys-terrorism-law.

[21] RSF (2021), New Greek law against disinformation endangers press freedom, Reporters Without Borders, https://rsf.org/en/news/new-greek-law-against-disinformation-endangers-press-freedom.

[179] RSF (2021), RSF Barometer, Reporters Without Borders, https://rsf.org/en/barometer?exaction_victimes_pays%5B113%5D=113&annee_start=2021&annee_end=2021&type%5Btue%5D=tue&&&#exaction-victimes.

[40] RSF (2020), “2020 World Press Freedom Index: “Entering a decisive decade for journalism, exacerbated by coronavirus””, Reporters Without Borders, https://rsf.org/en/2020-world-press-freedom-index-entering-decisive-decade-journalism-exacerbated-coronavirus (accessed on 29 November 2021).

[24] RSF (2020), “RSF’s 2020 Round-up: 50 journalists killed, two-thirds in countries “at peace””, Reporters Without Borders, https://rsf.org/en/news/rsfs-2020-round-50-journalists-killed-two-thirds-countries-peace.

[41] RSF (2020), “RSF’s 2020 Round-up: 50 journalists killed, two-thirds in countries “at peace””, Reporters Without Borders, https://rsf.org/en/news/rsfs-2020-round-50-journalists-killed-two-thirds-countries-peace (accessed on 29 November 2021).

[121] Šimonović, D. (2018), Report of the Special Rapporteur on violence against women, its causes and consequences on online violence against women and girls from a human rights perspective, UN Special Rapporteur on violence against women, https://undocs.org/A/HRC/38/47.

[100] Slovak Ministry of Interior (2021), Committee on the Prevention and Elimination of Racism, https://www.minv.sk/?pracovne-skupiny-vyboru.

[93] Slovak Ministry of the Interior (2020), The concept of the fight against radicalization and extremism until 2024, https://www.slov-lex.sk/legislativne-procesy/SK/LP/2020/582.

[104] Slovak National Centre for Human Rights (2019), 2019 Report of the Slovak national centre for human rights, http://www.snslp.sk/wp-content/uploads/Rocna-zaverecna-sprava-o-cinosti-SNSLP-za-rok-2019.pdf.

[90] Smith, A. (2018), Public Attitudes Toward Computer Algorithms, Pew Research Center, https://www.pewresearch.org/internet/2018/11/16/public-attitudes-toward-computer-algorithms/.

[92] Spanish Ministry of Interior (2019), Plan de Acción contra los delitos e incidentes de odio, http://www.interior.gob.es/documents/642012/3479677/plan+de+accion+delitos+de+odio/d054f47a-70f3-4748-986b-264a93187521.

[36] Stapenhurst, R. (2000), The Media Role in Curbing Corruption, World Bank, https://www.researchgate.net/publication/265232533_The_Media_Role_in_Curbing_Corruption (accessed on 18 May 2022).

[162] Ubaldi, B. et al. (2019), “State of the art in the use of emerging technologies in the public sector”, OECD Working Papers on Public Governance, No. 31, OECD Publishing, Paris, https://doi.org/10.1787/932780bc-en.

[14] UK Government (2021), Government publishes first ever national action plan to protect journalists, UK Government, https://www.gov.uk/government/news/government-publishes-first-ever-national-action-plan-to-protect-journalists (accessed on 13 September 2021).

[20] UN (2022), Türkiye: Concern mounts over human rights implications of ‘disinformation bill’, United Nations, https://news.un.org/en/story/2022/10/1129567.

[10] UN (2021), The promotion, protection and enjoyment of human rights on the Internet : resolution, United Nations Human Rights Council, https://digitallibrary.un.org/record/3937534?ln=en.

[70] UN (2020), Guidance Note on Addressing and Countering COVID-19 related Hate Speech, United Nations, https://www.un.org/en/genocideprevention/documents/Guidance%20on%20COVID-19%20related%20Hate%20Speech.pdf.

[165] UN (2020), Resolution adopted by the General Assembly on 16 December 2020, United Nations General Assembly, https://digitallibrary.un.org/record/3896430?ln=en.

[73] UN (2019), United Nations Strategy and Plan of Action on Hate Speech, United Nations, https://www.un.org/en/genocideprevention/hate-speech-strategy.shtml.

[42] UN (2018), The promotion, protection and enjoyment of human rights on, United Nations General Assembly, UN Human Rights Council, Geneva.

[46] UN (2016), Resolution 32/13 on the promotion, protection and enjoyment of human rights, United Nations Human Rights Council, https://daccess-ods.un.org/tmp/5330266.3564682.html.

[181] UN (2011), Council’s Resolution 16/18 on combating intolerance, negative stereotyping and stigmatization of, and discrimination, incitement to violence and violence against, persons based on religion or belief, United Nations Human Rights Council, https://www2.ohchr.org/english/bodies/hrcouncil/docs/16session/A.HRC.RES.16.18_en.pdf.

[120] UN Human Rights Committee (2021), “Concluding observations on the seventh periodic report of Germany”, https://tbinternet.ohchr.org/Treaties/CCPR/Shared%20Documents/DEU/CCPR_C_DEU_CO_7_47161_E.pdf (accessed on 31 May 2022).

[9] UN Human Rights Committee (2012), General comment No. 34, https://www2.ohchr.org/english/bodies/hrc/docs/gc34.pdf.

[22] UN Human Rights Council (2020), Resolution on the safety of journalists, https://digitallibrary.un.org/record/3888335 (accessed on 15 December 2021).

[50] UN et al. (2011), Joint Declaration on Freedom of Expression and the Internet, United Nations/Organization for Security and Co-operation in Europe/Organization of American States/African Commission on Human and Peoples’ Rights, https://www.osce.org/fom/78309.

[13] UNESCO (2021), Recent convictions highlight the work of Mexico’s Prosecutor Office dedicated to crimes against freedom of expression, https://en.unesco.org/news/recent-convictions-highlight-work-mexicos-prosecutor-office-dedicated-crimes-against-freedom (accessed on 15 December 2021).

[25] UNESCO (2021), UNESCO observatory of killed journalists, United Nations Educational, Scientific and Cultural Organization, https://en.unesco.org/themes/safety-journalists/observatory?field_journalists_date_killed_value%5Bmin%5D%5Byear%5D=2022&field_journalists_date_killed_value%5Bmax%5D%5Byear%5D=2022&field_journalists_gender_value_i18n=All&field_journalists_nationality_tid_i.

[31] UNESCO (2020), Safety of journalists covering protests : Preserving freedom of the press during times of turmoil, UNESCO Digital Library, https://unesdoc.unesco.org/ark:/48223/pf0000374206 (accessed on 28 February 2022).

[124] UNESCO (2015), Final report of the Broadband Commission Working Group on Gender, https://en.unesco.org/sites/default/files/highlightdocumentenglish.pdf.

[72] UNHCHR (2013), Annual report of the United Nations High Commissioner for Human Rights, https://www.ohchr.org/Documents/Issues/Opinion/SeminarRabat/Rabat_draft_outcome.pdf.

[108] Unia (2021), Stop Racisme, Cellule egalité des chances, Centre interfédéral pour l’égalité des chances, https://www.stopracisme.be/.

[35] V-Dem Institute (2021), Variables of Democracy, https://www.v-dem.net/en/analysis/VariableGraph/.

[135] Witchel, E. (2017), CPJ’s 2017 Global Impunity Index, https://cpj.org/reports/2017/10/impunity-index-getting-away-with-murder-killed-justice-2/.

[101] YWCA Canada (2020), YWCA Canada, https://ywcacanada.ca/what-we-do/projects-initiatives/block-hate-building-resilience-against-online-hate-speech/.

[109] Zivile Helden (2021), Beratungsstelle Hass im netz, https://www.zivile-helden.de/kontakt-beratung-fuer-zivile-helden/beratungsstellen-hass-im-netz/.

[6] Zuboff, S. (2019), The Age of Surveillance Capitalism: The Fight for a Human Future at the New Frontier of Power, Public Affairs, https://www.publicaffairsbooks.com/titles/shoshana-zuboff/the-age-of-surveillance-capitalism/9781610395694/.

Notes

← 1. ^. Unless otherwise stated, in line with the OECD Survey on Open Government, for the purposes of this report, the term citizen is meant in the sense of an inhabitant of a particular place and not as a legally recognised national of a state.

← 2. ^. This definition implicitly includes online and offline civic space. Notably, the definition of civic space adopted by the Development Assistance Committee (DAC) in 2021 in the DAC Recommendation on Enabling Civil Society in Development Co-operation and Humanitarian Assistance, adopted after the 2020 Survey on Open Government, includes an explicit mention of the virtual world: civic space is defined as the physical, virtual, legal, regulatory and policy space where people can, among other things, securely exercise their rights to the freedoms of peaceful assembly, association and expression, in keeping with human rights (OECD, 2021[182]).

← 3. ^. Armenia (Constitution Art. 42), Guatemala (Constitution Art. 35), Ireland (Constitution Art. 40) and Slovenia (Constitution Art. 39) are kindly asked to validate that freedom of the press is legally established.

← 4. ^. Reporters Without Borders (RSF) has, however, criticised the new legislation in Germany as not going far enough to protect the rights of journalists, as the amended version still allows the German intelligence service to collect, analyse and pass on the traffic data of media professionals and their contacts to other intelligence agencies without restrictions. See RSF (2021[177]).

← 5. ^. See Rice v. Vice Media Canada, Inc., Judgment of 30 November 2018.

← 6. ^. The OECD defines misinformation as false or inaccurate information not disseminated with the intention of deceiving the public and disinformation as false, inaccurate or misleading information deliberately created, presented and disseminated to deceive the public.

← 7. ^. The number of killed journalists reported by different organisations in 2021 differs due to the use of different methodologies. The UNESCO observatory of killed journalists reports 55 killings (39 for 2022) (2021[25]); the Committee to Protect Journalists reports 45 killings (17 with an unconfirmed motive; 28 with a confirmed motive; 38 in total for 2022) (CPJ, 2021[29]); the RSF reports 31 killings (29 journalists; 2 media workers; same number for 2022) (RSF, 2021[179]); and the International Federation of Journalists reports 45 killings (IFJ, 2021[180]).

← 8. ^. The V-Dem Institute’s indicator on harassment of journalists is based on the evaluation of multiple ratings provided by country experts with deep knowledge of a country and of a particular political institution, of whom about 85% are academics or professionals working in media or public affairs (e.g. senior analysts, editors, judges); about two-thirds are also nationals of and/or residents in a country and have documented knowledge of both that country and a specific substantive area. The question related to this indicator is: “Are individual journalists harassed – i.e. threatened with libel, arrested, imprisoned, beaten, or killed – by governmental or powerful non-governmental actors while engaged in legitimate journalistic activities?”. Responses include the following options:

0: No journalists dare to engage in journalistic activities that would offend powerful actors because harassment or worse would be certain to occur.
1: Some journalists occasionally offend powerful actors but they are almost always harassed or worse and eventually are forced to stop.
2: Some journalists who offend powerful actors are forced to stop but others manage to continue practising journalism freely for long periods of time.
3: It is rare for any journalist to be harassed for offending powerful actors, and if this were to happen, those responsible for the harassment would be identified and punished.
4: Journalists are never harassed by governmental or powerful non-governmental actors while engaged in legitimate journalistic activities.

Given that the data of the V-Dem Institute’s indicator on harassment of journalists for 2021 did not

← 9. ^. For the purposes of the 2020 OECD Survey on Open Government, an open Internet was defined as follows: a fundamental network neutrality concept in which information across the World Wide Web is equally free and available without variables that depend on the financial motives of Internet service providers.

← 10. ^.See the Council of Europe Recommendation CM/Rec(2011)8 of the Committee of Ministers to member states on the protection and promotion of the universality, integrity and openness of the Internet, adopted by the Committee of Ministers on 21 September 2011 at the 1 121^st meeting of the Ministers’ Deputies, para. 11. For more information: https://rm.coe.int/CoERMPublicCommonSearchServices/DisplayDCTMContent?documentId=0900001680645b44.

← 11. ^.Net neutrality refers to the principle that Internet service providers should enable access to all content and applications regardless of the source, and without favouring or blocking particular products or websites.

← 12. ^. The OECD defines misinformation as false or inaccurate information not disseminated with the intention of deceiving the public.

← 13. ^. The V-Dem Institute’s indicator on Internet filtering in practice is based on the evaluation of multiple ratings provided by country experts with deep knowledge of a country. The question related to this indicator is: “How frequently does the government censor political information (text, audio, images, or video) on the Internet by filtering (blocking access to certain websites)?”. Responses include the following:

0: Extremely often. It is a regular practice for the government to remove political content, except to sites that are pro-government.
1: Often. The government commonly removes online political content, except sites that are pro-government.
2: Sometimes. The government successfully removes about half of the critical online political content.
3: Rarely. There have been only a few occasions on which the government removed political content.
4: Never, or almost never. The government allows Internet access that is unrestricted, with the exceptions mentioned in the clarifications section.

← 14. ^. As discussed in Chapter 2, the UN launched a Strategy and Plan of Action on Hate Speech in 2013 (UNHCHR, 2013[72]). The Istanbul Process – a group of countries that united around the UN Human Rights Council’s Resolution 16/18 on combatting intolerance, negative stereotyping and stigmatisation of, and discrimination, incitement to violence and violence against, persons based on religion or belief, was also reinvigorated in 2019, working to promote tolerance and inclusion, and end violence and discrimination based on religion or belief (UN, 2011[181]).

← 15. ^. See European Court of Human Rights, Delfi AS v. Estonia, Application No. 64569/09, [GC] Judgment of 16 June 2015, para. 162. By contrast, in the absence of hate speech or any direct threats to physical integrity in the user comments in question, the court has found that objective liability of Internet portals for third-party comments was not compatible with Article 10 of the convention. See Magyar Tartalomszolgáltatók Egyesülete and Index.hu Zrt v. Hungary, Application No. 22947/13, Judgment of 2 February 2016, para. 91, and Savva Terentyev v. Russia, Application No. 10692/09, Judgment of 28 August 2018, paras. 83-87, where the court found that a blogger’s offensive statements could not be regarded as an attempt to incite hatred or provoke violence, and thus concluded that his criminal conviction was not proportionate.

← 16. ^. For a list of the Council of Europe member states’ reporting mechanisms for online hate speech, see CoE (2021[175]).

← 17. ^. In 2014, a global survey on harassment and violence against female media workers launched in August 2013 and completed by almost 1 000 women from around the world – conducted by the International Women’s Media Foundation (IWMF) and the International News Safety Institute (INSI) – found that 23% of women respondents had experienced intimidation, threats or abuse online in relation to their work (Barton, 2014[178]). A follow-up survey conducted by the IWMF and Trollbusters in 2018 found that 63% of women respondents had been harassed or abused online (Ferrier, 2018[131]). In 2021, a UNESCO- International Center for Journalists survey found that 73% of women journalists had experienced online violence (Posetti, 2021[123]). While these surveys are not directly comparable, the pattern suggests that gendered online violence against women journalists has worsened significantly over the past few years.

← 18. ^. The OECD defines misinformation as false or inaccurate information not disseminated with the intention of deceiving the public and disinformation as false, inaccurate or misleading information deliberately created, presented and disseminated to deceive the public.

← 19. ^. The OECD has adopted the following definition of artificial intelligence: AI refers to a machine-based system that can, for a given set of human-defined objectives, make predictions, recommendations or decisions influencing real or virtual environments. AI systems are designed to operate with varying levels of autonomy.

← 20. ^.AI is used for predictive policing in a number of OECD Members, including Belgium, Germany, the Netherlands, Spain and the United Kingdom (Allen and Masters, 2020[176]).

← 21. ^. According to a 2019 Eurobarometer survey, only 40% of Europeans know that they can have a say when decisions are automated. See https://search.gesis.org/research_data/ZA7562?doi=10.4232/1.13318

╳

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

https://doi.org/10.1787/d234e975-en

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at https://www.oecd.org/termsandconditions.