CNTI offers practical and accessible digital resources to promote evidence-based global discussions about the challenges facing an independent press and an open internet.

This resource includes aggregated demographic, media and technology data for 179 countries and markets from independent global institutions such as the United Nations, the World Bank, V-Dem Institute, Reporters Without Borders, the Reuters Institute for the Study of Journalism, the World Justice Project, the Committee to Protect Journalists and Freedom House. The first dataset is in tabular form, and the mapped version with selected data is below it.

¹The World Bank classifies economies into four income groups: low, lower-middle, upper-middle, and high income. For this purpose it uses gross national income (GNI) per capita data in U.S. dollars, converted from local currency.

²V-Dem relies on expert assessments to annually classify four regime types — liberal democracy, electoral democracy, electoral autocracy and closed autocracy — based on the Regimes of the World measure. V-Dem incorporates confidence estimates in order to account for the uncertainty and potential measurement error due to the nature of the data but also to underline that some countries are placed in the grey zone between regime types: (+) indicates signifies that taking uncertainty into account, the country could also belong to the higher category, while (–) indicates that taking uncertainty into account, the country could belong to the lower category. For more information on scoring and weighting, see their codebook and methodology.

³The World Justice Project relies on expert assessments to annually evaluate the rule of law, drawing on nine factors and 47 sub-factors, in 142 countries and jurisdictions. The rule of law is a durable system of laws, institutions, norms, and community commitment that delivers accountability, just law, open government and accessible and impartial justice. Responses are mapped onto the 44 sub-factors of the Index, codified so all values fall between 0 (weakest adherence to the rule of law) and 1 (strongest adherence to the rule of law), and aggregated at the country level using the simple (or unweighted) average of all respondents. For more information on scoring and weighting, see their methodology.

⁴Freedom House relies on expert assessments to annually compare the level of internet freedom in 70 countries around the world. Scores are based on a scale of 0 (least free) to 100 (most free). Countries are then ranked from 1-70. For more information on scoring and weighting, see their methodology. Note: Research has found that measures of internet censorship relying on expert analysis are less likely to include false negatives (Fletcher and Hayes-Birchler, 2023). In other words, as many instances of censorship are captured as possible, potentially including some countries that may not have engaged in censorship. Because some governments try to conceal censorship measures, proponents of this method argue that remote measurement methods may miss more nuanced cases of censorship.

⁵V-Dem relies on expert assessments to annually assess to what extent governments attempt to censor information (text, audio, or visuals) on the internet. Responses are assessed on a 0-3 scale, where 0 represents “The government successfully blocks Internet access except to sites that are pro-government or devoid of political content” and 3 represents “The government allows Internet access that is unrestricted.” For ease of interpretation, interval-level point estimates have been linearly transformed back to the 0-3 coding scale, and the heat map represents linear quartiles (from light yellow for the bottom quartile to dark green for the top quartile). For more information on scoring and weighting, see their codebook and methodology.

⁶The Tortoise Global AI Index annually ranks countries based on capacity for artificial intelligence, by measuring levels of investment, innovation and implementation. The GAII draws on a range of 28 different data sources and 111 indicators. For more information on scoring and weighting, see their methodology.

⁷V-Dem relies on expert assessments to annually assess to what extent governments respect press and media freedom, the freedom of ordinary people to discuss political matters at home and in the public sphere and the freedom of academic and cultural expression. The index is aggregated on a 0-1 scale and is formed by taking the point estimates from a Bayesian factor analysis model of the indicators for media censorship effort, harassment of journalists, media bias, media self-censorship, print/broadcast media criticism, print/broadcast media perspectives, freedom of discussion for men/women and freedom of academic and cultural expression. For more information on scoring and weighting, see their codebook and methodology.

⁸V-Dem relies on expert assessments to annually assess to what extent governments attempt to censor press and broadcast media. Responses are assessed on a 0-4 scale, where 0 represents “Attempts to censor are direct and routine” and 4 represents “The government rarely attempts to censor major media in any way, and when such exceptional attempts are discovered, the responsible officials are usually punished.” For ease of interpretation, interval-level point estimates have been linearly transformed back to the 0-4 coding scale, and the heat map represents linear quartiles (from light yellow for the bottom quartile to dark green for the top quartile). For more information on scoring and weighting, see their codebook and methodology.

⁹The RSF World Press Freedom Index relies on expert assessments to annually compare the level of freedom enjoyed by journalists and media in 180 countries and territories, based on five distinct categories or indicators (political context, legal framework, economic context, sociocultural context and safety). The Index is based on a score ranging from 0 to 100 that is assigned to each country or territory, with 100 being the best possible score (the highest possible level of press freedom) and 0 the worst. Countries are then ranked from 1-180. For more information on scoring and weighting, see their methodology.

¹⁰The Reuters Institute for the Study of Journalism annually measures overall media trust based on the percentages of respondents in each market who say they ‘agree’ in response to the question: “Thinking about news in general, do you agree or disagree with the following statements? – I think you can trust most news most of the time.” For more information on samples and weighting, see their methodology.

¹¹V-Dem relies on expert assessments to annually assess whether individual journalists are harassed — i.e., threatened with libel, arrested, imprisoned, beaten, or killed — by governmental or powerful nongovernmental actors while engaged in legitimate journalistic activities. Responses are assessed on a 0-4 scale, where 0 represents “No journalists dare to engage in journalistic activities that would offend powerful actors because harassment or worse would be certain to occur” and 4 represents “Journalists are never harassed by governmental or powerful nongovernmental actors while engaged in legitimate journalistic activities.” For ease of interpretation, interval-level point estimates have been linearly transformed back to the 0-4 coding scale, and the heat map represents linear quartiles (from light yellow for the bottom quartile to dark green for the top quartile). For more information on scoring and weighting, see their codebook and methodology.

¹²The Committee to Protect Journalists regularly tracks attacks on the press, including imprisonment and killings. These two trackers are merged here for an overall count of imprisonments and killings of journalists in the most recent (completed) calendar year. For more information on this tracker, see their methodology.

The post Aggregated Country Data appeared first on Center for News, Technology & Innovation.

Artificial Intelligence in Journalism

In recent years, a range of legislative proposals have been introduced that would, to some extent, regulate AI. The AI Bill of Rights, unveiled in October 2022, outlines five key protections Americans should have in the AI age upon which to build future policy discussions to address safety, discrimination, privacy and transparency. This document may set the stage for future legislation or for the establishment of an agency to regulate AI but lacks details and mechanisms for AI regulation and enforcement. The soon-to-be-reintroduced collective bargaining legislation, the Journalism Competition & Preservation Act (JCPA), is also expected to include commercial arrangements related to AI tools. In late March 2023, the U.S. Copyright Office launched an initiative to examine AI issues, including “the use of copyrighted materials in AI training.” In October 2023, the Biden administration released an executive order to develop safety guidelines around generative AI. The Institute for Data, Democracy & Politics maintains a list of U.S. federal legislative proposals pertaining to generative AI, and other organizations have tracked this legislation at the state level.

Building News Sustainability

The Journalism Competition and Preservation Act (JCPA), dropped by U.S. Congress in December 2022 but revived and advanced by the Senate Judiciary Committee in June 2023, would create an exemption to antitrust law allowing news publishers to band together to negotiate for payments from digital platforms. Concerns have been expressed about the bill benefitting legacy or large publishers at the expense of local, independent media (e.g., a publisher cap of 1,500 employees excluding only the three largest U.S. newspapers), its potential to threaten platforms’ efforts to moderate hate speech and disinformation, its potential to violate existing copyright law and its missing requirements that publishers spend money on journalism (e.g., hiring journalists or boosting coverage of news).

Similar legislation is also being proposed at the state level. For instance, in Massachusetts, policymakers are considering proposed legislation, H.74 and S.34, to collect fees from streaming service companies into a general fund, with a portion of the fees earmarked specifically to support public access media. Further, the California Journalism Preservation Act, put on hold in the state’s legislature until 2024, would tax platforms’ advertising profits made from distributing news articles as a “usage fee” and redistribute at least 70% of that money to news publishers based on how often their content is displayed. Concerns have been expressed about the bill’s potential to violate the First Amendment and existing copyright law, harm smaller publishers and financially compensate creators of clickbait and disinformation. Questions remain about how the bill would be implemented across digital platforms that operate differently (e.g., social media versus search engines) and about how to ensure funding would go toward journalism. Like in Canada, in response to the passage of the CJPA, platform companies including Meta and Google have announced that they will block news links on their platforms.

Separately, recent U.S. legislation focuses on economic support for local news specifically with the revamped Community News and Small Business Support Act, which was reintroduced in July 2023 to the U.S. House Committee on Ways and Means. The bill would create an advertising tax credit for small businesses to advertise with local news and a payroll tax credit for local journalists, both for five years. It also includes a provision preventing (1) news organizations controlled or funded by interest groups and (2) newsrooms that don’t employ at least one full-time local journalist from eligibility. An earlier iteration of this bill, the Local Journalism Sustainability Act, floundered in the U.S. House of Representatives in 2021, in part due to its inclusion of a complicated third individual tax credit for news subscriptions.

Enhancing Algorithmic Transparency

The U.S.’s most recent platform accountability legislation, the Platform Accountability and Transparency Act, was reintroduced to Congress in June 2023. One central element of the act is its public transparency requirements for social media platforms, including provisions for public access to advertising libraries and disclosures on “highly disseminated,’’ or viral, content. 

The act (along with the Digital Services Oversight and Safety Act) would also promote and protect researcher access to platform data by creating a National Science Foundation (NSF)-facilitated program for independent researchers to request access to certain data, requiring the development of privacy and security protocols for the program and offering legal protections for researchers and journalists who collect and analyze publicly available platform data. A provision to revoke controversial Section 230 protections from platforms that fail to comply with approved data requests was removed from the bill. 

Separately, the Biden administration may seek a deal with European regulators to harmonize access for U.S. researchers with provisions in the EU’s Digital Services Act (DSA).

Journalists & Digital Security Threats

In early 2023, the Biden administration released a wide-ranging cybersecurity strategy aiming to bolster protections against cyberattacks. Separately, the administration has taken action to counter spyware abuses, including announcing a multi-country collaboration to control the proliferation of these technologies and signing an executive order to prohibit U.S. governmental use of commercial spyware.

Meanwhile, bipartisan U.S. House and Senate bills (the PRESS Act) have been reintroduced with the aim of shielding journalists’ communication records and data. The act would limit the government’s ability to require journalists to disclose data that could identify their sources, with limited exceptions for terrorism and the imminent threat of violence. The PRESS Act has largely been lauded by experts as creating strong protections for independent journalism.

At the same time, U.S. lawmakers have also been considering online child-safety bills such as the EARN It Act and the Stop CSAM Act, but there are concerns that they could be exploited to spy on users’ private conversations, thus eroding encryption protections and threatening human rights.

Protecting an Open Internet

In April 2022, the Biden administration announced a global partnership signed by 61 nations, the Declaration for the Future of the Internet, to establish norms for an open, global internet. While the declaration is largely symbolic, and signatories include countries with declining or inconsistent internet freedoms, some experts note it may help to prevent wavering democracies from state overreach when it comes to internet governance. 

Within the same month, the U.S. Department of the Treasury also announced an order allowing internet service providers to continue operating in Russia despite sanctions following the country’s invasion of Ukraine. The order came in response to warnings from civil society groups that restricting internet access by pulling U.S.-based software companies and internet providers from the country could risk further isolating Russian citizens and journalists. The order promoted an open flow of information for citizens and independent media.

At the same time, new challenges to an open internet in the U.S. have emerged as some policymakers have called for bans of the popular platform TikTok. In 2020, the Trump administration issued a confusing series of executive orders (later challenged in court and revoked by the Biden administration in 2021) intended to block apps like TikTok, WeChat and Alipay due to national security threats without offering clear evidence of the nature of these threats. In 2022 and 2023, several legislative proposals restricting TikTok have been introduced at both the federal and the state level. In particular, Montana’s legislation outright banning TikTok, S.B. 419, targets app stores and TikTok, rather than individuals, imposing a daily fine for each time a user is given the ability to access TikTok. The proposal, if not vetoed by Montana’s governor, would take effect in January 2024. The ban has since been challenged in lawsuits by TikTok users and by TikTok itself. As we note earlier in this primer, these bans represent threats to free expression and an open internet.

The post Country in Focus: United States appeared first on Center for News, Technology & Innovation.

Artificial Intelligence in Journalism

The EU’s law on artificial intelligence, the AI Act, is arguably the most developed law about AI and includes regulations that assign the application of AI to three risk categories. The law bans uses of generative AI systems with “unacceptable” risk levels (i.e., AI systems that violate people’s fundamental rights and safety) and increases regulation of “high-risk” systems. Analyses of whether foundation model providers currently comply with these requirements have found they largely do not. 

There is some discussion about the EU AI Act eventually serving as a global standard for policy. However, concerns about the AI Act include a lack of clarity of what constitutes AI, a lack of flexibility in the regulation process and unintended legal implications for marginalized communities (e.g., migrants and refugees). The European Parliament voted to adopt the act in June 2023. Updates on the latest status can be found on the AI Act website. 

Enhancing Algorithmic Transparency

Policymakers in the EU have begun addressing algorithmic transparency as part of the broader Digital Services Act (DSA), which entered into force in November 2022. In particular, the act requires platforms to provide more transparency around their content moderation and algorithmic content curation efforts, to take further steps to address harmful content and to disclose data to independent researchers. In their terms of service, very large platforms must also publish the primary parameters of their recommendation systems. In addition, the DSA asks companies to give users the right to opt out of recommendation algorithms based on personal data processing and instead engage organically with content not based on profiling. Concerns about the act often focus on its feasibility on a cross-national scale: how regulators will monitor these standards and whether individual countries may follow through with such regulation.

Protecting an Open Internet

The European Union’s sweeping 2018 data protection legislation, known as the GDPR, created new means of safeguarding citizens’ personal data but in doing so introduced new concerns around  a “splinternet” experience, where one’s location would determine what content is banned. In particular, the GDPR’s ‘right to be forgotten’ risks fact-based investigative journalism including personal data (such as #MeToo stories) qualifying for removal demands. (Between 2015 and 2021, Google and Bing received over 1 million ‘right to be forgotten’ requests.)

Like the United States, the European Union introduced exemptions in April 2022 for internet service providers to continue operating in Russia despite sanctions following the country’s invasion of Ukraine. The order came in response to warnings from civil society groups that restricting internet access by pulling EU-based software companies and internet providers from the country could risk further isolating Russian citizens and journalists. The order promoted an open flow of information for citizens and independent media.

Meanwhile, the EU’s 2022 ban of Russian state media RT and Sputnik due to legitimate concerns about the spread of propaganda and disinformation, raised important questions surrounding the risk of closing of the global information space and cutting off the ability of policymakers, journalists, activists and citizens to learn how state media may try to mislead or manipulate public opinion. Additionally, some experts question whether it sets a precedent for further legal media bans in Europe.

The post Countries in Focus: European Union appeared first on Center for News, Technology & Innovation.

Building News Sustainability

Australia’s News Media Bargaining Code, passed into law by Parliament in February 2021, was the first regulatory legislation of its kind to come into effect and has since served as a model for policymaking in countries like Canada and Brazil. The code included a late provision that platforms could avoid “designation,” or being subject to arbitration, if they made a “significant contribution” to the sustainability of Australia’s news industry. While the code faced strong resistance from major platform companies – with Google threatening to cut off search and Facebook briefly blocking news feeds – at least 30 commercial agreements have since been made to compensate dozens of outlets (though these deals are far from transparent). The chief criticisms of the Australian code include that it primarily benefits legacy or large publishers at the expense of local, independent media; it does not include a transparency requirement, so the size of deals with publishers is unknown; and outlets are not required to spend money on journalism (e.g., hiring journalists or boosting coverage of news). Of particular concern is what has been deemed the “poison pill” clause: that if platforms become “designated,” their deals with publishers become void.

The post Country in Focus: Australia appeared first on Center for News, Technology & Innovation.

Addressing Disinformation

Brazilian Bill PL 2630 – known as the “Fake News” bill – narrowly failed a fast-track vote in April 2022. The bill originally stalled after technology companies and independent fact-checking organizations expressed concern about the bill’s lack of a clear definition of “journalism.” Drawing from Australia’s News Media Bargaining Code, PL 2630 would require platforms to pay for journalistic content, enhance transparency requirements and detect and remove illegal content.

In the wake of antidemocratic attacks and political violence connected to disinformation and inflammatory speech on social platforms in Brazil, a new vote on PL 2630 was scheduled for early May 2023 and similarly stalled after an influx of last-minute amendments. Concerns were expressed about unintended consequences of the bill, including its threat to free expression and potential to financially compensate creators of disinformation. Aggressive platform responses to shape debate over the law received international criticism, and experts noted concerns with the escalation of responses and its impact on free expression.

Artificial Intelligence in Journalism

In September 2021, Brazil’s Chamber of Deputies approved the Brazilian Legal Framework for Artificial Intelligence (Marco Legal da Inteligência Artificial) to regulate the development and use of AI technologies and to promote research on AI ethics and accountability. Experts have expressed concerns that the legislation does not thoroughly address AI accountability and transparency, but rather may make accountability for AI discriminatory biases harder to achieve. Further, given the complexity and autonomy of AI systems, experts have noted the legal difficulty of what Brazilian law calls “subjective liability,” which assigns the responsibility of proving algorithmic errors to the individuals seeking damages. 

In May 2023, Brazilian Senate Speaker Rodrigo Pacheco Thursday submitted a bill to regulate AI use based on recommendations made by a working group created in 2022. The stated aim of this legislation is to protect citizens’ fundamental rights and, similar to the EU AI Bill of Rights, introduces a risk-based regulatory model for AI systems.

Building News Sustainability

The Brazilian government proposed a media bargaining code as part of a larger set of digital platform regulations. Bill PL 2630 – known as the “Fake News” bill – narrowly failed a fast-track vote in April 2022 and stalled again in May 2023 after an influx of last-minute amendments. PL 2630 would recognize publishers’ rights to be paid for journalistic content – drawing from Australia’s News Media Bargaining Code – as well as enhance transparency requirements and establish a broad range of other obligations. Brazil’s law, like Australia’s, did not include transparency requirements for its bargaining structure, and concerns were expressed by the private sector about the bargaining code’s potential to financially compensate creators of disinformation. Aggressive platform responses to shape debate over the law received international criticism, and experts noted concerns with the escalation of responses and its impact on free expression.

The post Country in Focus: Brazil appeared first on Center for News, Technology & Innovation.

Algorithms & Quality News

Canadian Parliament’s new law, Bill C-18, requires digital platforms to compensate media outlets for reposting or linking to their content, with the stated aim of providing “access to quality, fact-based news at the local and national levels.” Several elements of this bill pertain directly to the challenge of determining and assessing “quality” journalism. Without a clearly established basis for payments, platforms’ use of content reach or volume as key metrics for payments may end up incentivizing cheap, low-quality content. An “undue or unreasonable preference” provision intended to protect publishers from platform retaliation may also impact existing algorithmic structures prioritizing fact-based content and original reporting. Finally, the bill requires publishers to meet standards to be eligible for bargaining set by the Canadian Radio-television and Telecommunications Commission – which may serve as a template for governmental bodies to serve as the arbiters of content.

Building News Sustainability

In June 2023, Canada passed its Online News Act, Bill C-18, to go into effect in December 2023. Using Australia’s News Media Bargaining Code as its starting point, C-18 narrowed the platform exemption process (with non-designated platforms still subject to some regulation) and added several transparency requirements, including participating organizations needing to report deals to a regulator and an amendment ensuring some (though not a majority of) funding goes to newsroom staffing. Concerns about the bill include the ambiguity of its inclusion criteria potentially harming smaller publishers, its “undue privilege” clause (while aiming to prevent retaliation against publishers) potentially harm platforms’ ability to promote high-quality content or curb disinformation and its potential knock-on effect of further commercializing the internet overall. Googlehas agreed to pay $100M CAD annually in financial support to the Canadian news industry, whereas platforms like Meta have blocked Canadian news links in response to the legislation. 

This is not the first initiative of Canada’s to provide policy-driven economic support for news. In 2020, the Canadian government developed the Local Journalism Initiative, a $50 million, five-year program funding local and civic journalism for underserved communities. The initiative raised critical questions about how to determine who receives such support.

The post Country in Focus: Canada appeared first on Center for News, Technology & Innovation.

Protecting an Open Internet

India’s government has expanded its digital censorship efforts since 2019, including through legislation, internet shutdowns, platform bans and heightened scrutiny of platforms. The Indian government abruptly banned TikTok in June 2020, citing issues of privacy and national sovereignty. In 2021, an executive order introduced sweeping internet regulations forcing technology companies to comply with government surveillance and censorship requirements and undermining user rights. In particular, it targeted end-to-end encryption protocols used by platforms like WhatsApp, requiring messaging apps to trace a message’s “first originator” to reveal the identities of senders. As we note earlier in this primer, these bans threaten free expression and an open internet.

In 2022, India’s cybersecurity agency (CERT-In) implemented new regulations on national security grounds requiring the retention of customer data by VPN, web and cloud service providers for five years, leading some VPN providers to announce withdrawal of their services in the country. This system restricts a free flow of information between India and the rest of the world.

In 2023, the Indian government issued legal notices pressuring Twitter to block accounts critical of Prime Minister Narendra Modi’s government based on claims that they violated India’s Information Technology Act, making the accounts inaccessible in India. New regulations were also enacted in April 2023 requiring platforms to remove content deemed false or misleading about the Indian government by a Ministry of Electronics and Information Technology-appointed body.

The post Country in Focus: India appeared first on Center for News, Technology & Innovation.