Disinformation affecting the EU: tackled but not tamed

Executive summary

Disinformation has been present in human communication since the dawn of civilisation and the creation of organised societies. What has changed in recent years, however, is its sheer scale and the speed with which false or misleading information can reach its intended and unintended audiences through social media and new technologies. This may cause public harm.

The European Council, in its conclusions of 28 June 2018, invited the EU’s High Representative for Foreign Affairs and Security Policy and the Commission to present an action plan with specific proposals for a coordinated response to disinformation. The EU action plan against disinformation, presented on 5 December 2018, includes ten specific actions based on four priority areas or ‘pillars’, and consolidates the EU’s efforts to tackle disinformation. EU spending on tackling disinformation has been relatively low to date: €50 million between 2015 and 2020.

The purpose of the audit was to assess whether the EU action plan against disinformation was relevant when drawn up and delivering its intended results. The audit covered the period from the run-up to the adoption of the EU action plan against disinformation in December 2018 until September 2020. This report is the first comprehensive, independent assessment of its relevance and the results achieved. Overall, we conclude that the EU action plan was relevant but incomplete, and even though its implementation is broadly on track and there is evidence of positive developments, some results have not been delivered as intended.

The EU action plan contains relevant, proactive and reactive measures to fight disinformation. However, even though disinformation tactics, actors and technology are constantly evolving, the EU action plan has not been updated since it was presented in 2018. It does not include comprehensive arrangements to ensure that any EU response against disinformation is well coordinated, effective and proportionate to the type and scale of the threat. Additionally, there was no monitoring, evaluation and reporting framework accompanying the EU action plan, which undermines accountability.

The European External Action Service’s three strategic communications task forces have improved the EU’s capacity to forecast and respond to disinformation in neighbouring countries. However, they are not adequately resourced or evaluated, and their mandates do not cover some emerging threats.

The EUvsDisinfo project has been instrumental in raising awareness about Russian disinformation. However, the fact that it is hosted by the European External Action Service raises some questions about its independence and ultimate purpose, as it could be perceived as representing the EU’s official position. The rapid alert system has facilitated information sharing among Member States and EU institutions. Nevertheless, Member States are not using the system to its full potential for coordinating joint responses to disinformation and common action.

With the code of practice, the Commission has established a pioneering framework for engagement with online platforms. We found that the code of practice fell short of its goal of holding online platforms accountable for their actions and their role in actively tackling disinformation.

The report also highlights the absence of a media literacy strategy that includes tackling disinformation, and the fragmentation of policy and actions to increase capacity to access, understand and interact with media and communications. Finally, we found there was a risk that the newly created European Digital Media Observatory would not achieve its objectives.

Based on these conclusions, we recommend that the European External Action Service and the Commission:

• improve the coordination and accountability of EU actions against disinformation (European External Action Service and Commission);
• improve the operational arrangements of the StratCom division and its task forces (European External Action Service);
• increase participation in the rapid alert system by Member States and online platforms (European External Action Service);
• improve the monitoring and accountability of online platforms (Commission);
• adopt an EU media literacy strategy that includes tackling disinformation (Commission);
• take steps to enable the European Digital Media Observatory to fulfil its ambitious objectives (Commission).

Introduction

The European Commission¹ defines “disinformation” as the “creation, presentation and dissemination of verifiably false or misleading information for the purposes of economic gain or intentionally deceiving the public, which may cause public harm”. Such public harm includes threats to democratic political and policy-making processes as well as to the protection of EU citizens' health, the environment or security.

The Commission’s definition of disinformation excludes misleading advertising, reporting errors, satire and parody, or clearly identified partisan news and commentary. Unlike hate speech or terrorist material, for example, false or misleading information is not illegal on its own.

The EU’s legitimacy and purpose rest on a democratic foundation, which depends on an informed electorate expressing its democratic will through free and fair elections. Any attempt to maliciously and intentionally undermine or manipulate public opinion therefore represents a grave threat to the EU itself. At the same time, combatting disinformation represents a major challenge, as it should not impair the freedom of opinion and expression enshrined in the EU Charter of Fundamental Rights.

The term “disinformation” emerged in the early 20^th century and has been used extensively ever since. In recent years, the internet has amplified the scale and speed with which false information is reaching audiences, often anonymously and at minimal cost.

EU efforts to combat disinformation date back to March 2015, when the European Council² invited the EU’s High Representative for Foreign Affairs and Security Policy (the “High Representative”), in cooperation with the Member States and EU institutions, “to develop an action plan on strategic communication to address Russia’s ongoing disinformation campaigns”. This led to the creation of the strategic communications division (“StratCom”) and the first of its task forces within the European External Action Service (EEAS), with a mandate to counter disinformation originating outside the EU (Russia) and design and disseminate positive strategic communications in the EU’s eastern neighbourhood – known as the East strategic communications task force. In 2017, two more StratCom task forces were created: one for the southern neighbourhood and another for the Western Balkans (see also paragraphs 45-49).

In late 2017, the Commission set up a high-level expert group to offer concrete advice on tackling disinformation. The group delivered its report in March 2018³, which formed the basis for the Commission’s “Communication on tackling online disinformation: a European approach”⁴ (April 2018). This Communication outlined key overarching principles and objectives to guide action to raise public awareness about disinformation, as well as the specific measures the Commission intended to take.

The European Council, in its conclusions of 28 June 2018⁵, invited the EU’s High Representative for Foreign Affairs and Security Policy and the Commission to “present an action plan by December 2018 with specific proposals for a coordinated EU response to the challenge of disinformation, including appropriate mandates and sufficient resources for the relevant EEAS Strategic Communications teams”.

Based on the April 2018 communication, the Commission published an EU action plan against disinformation in December 2018 (hereinafter: the “EU action plan”). It sets out ten specific actions based on four priority areas or ‘pillars’ targeting society as a whole, as shown in Table 1 below.

There is no EU legal framework governing disinformation apart from article 11 of the Charter on Fundamental Rights on the Freedom of expression and information, and a series of policy initiatives. Responsibility for combatting disinformation lies primarily with the Member States⁶. The EU's role is to support the Member States with a common vision and actions aimed at strengthening coordination, communication and the adoption of good practices. Annex I presents the main departments and offices of the EU institutions involved in the implementation of the EU action plan. As Annex II shows, EU spending on tackling disinformation has been relatively low to date: €50 million between 2015 and 2020.

In December 2019⁷, the Council confirmed that the EU action plan “remains at the heart of the EU’s efforts” to combat disinformation, and called for it to be reviewed regularly and updated where necessary. It also invited the EEAS to reinforce its strategic communication work in other regions, including sub-Saharan Africa. The European Parliament has also on numerous occasions expressed the importance of strengthening efforts to combat disinformation⁸.

In early 2020, almost immediately following the COVID‑19 outbreak, an unprecedented wave of misinformation, disinformation and digital hoaxes appeared on the internet, which the World Health Organization described as an “infodemic”⁹. This posed a direct threat to public health and economic recovery. In June 2020, the European Commission and the High Representative published a communication entitled “Tackling COVID‑19 disinformation - Getting the facts right”¹⁰, which looked at the steps already taken and concrete actions to follow against disinformation regarding COVID‑19.

On 4 December 2020, the Commission presented the “European democracy action plan”¹¹, part of which is dedicated to strengthening the fight against disinformation. It builds on existing initiatives in the EU action plan against disinformation. Furthermore, the Commission also issued a proposal for a Digital Services Act¹² proposing a horizontal framework for regulatory oversight, accountability and transparency of the online space in response to the emerging risks.

Figure 1 provides a timeline of the main EU initiatives since 2015.

Source: ECA.

Audit scope and approach

This audit report comes two years after the adoption of the EU action plan against disinformation. It is the first comprehensive, independent assessment of its relevance and the results achieved, thereby contributing to the regular review of the EU action plan requested by the Council.

The aim of our audit was to ascertain whether the EU action plan against disinformation is relevant and delivering its intended results. To answer this question, we addressed two sub-questions:

• Is the EU action plan relevant for tackling disinformation and underpinned by a sound accountability framework?
• Are the actions in the EU action plan being implemented as planned? To answer this sub-question, we assessed the status of the actions under each of the four pillars of the plan.

The audit covered the period from the run-up to the adoption of the EU action plan against disinformation in December 2018 until September 2020. Where possible, this report also refers to recent developments in this area after this date, such as the Commission’s presentation of the European democracy action plan and the proposal for a Digital Services Act in December 2020 (see paragraph 12). However, as these documents were published after we had completed our audit work, they are beyond the scope of this audit.

The audit included an extensive desk review and analysis of all available documentation on the structures put in place, and the actions planned and implemented through the EU action plan. We sent a survey to the 27 Member States’ Rapid Alert System contact points, with a response rate of 100 %. We also held meetings with numerous stakeholders such as the EEAS and relevant Commission directorates-general (DGs), the European Parliament, the Council, Commission representations, NATO, the NATO-affiliated Strategic Communication Centre of Excellence in Latvia, national authorities, online platforms, journalist and fact-checking organisations, audio-visual media services regulating bodies which provide advice to the Commission, academics and experts, project managers/coordinators and an external expert.

We also assessed 20 out of 23 projects the Commission identified as being directly related to fighting disinformation through media literacy. Annex III summarises our assessment of these projects.

Observations

The EU action plan against disinformation was relevant when drawn up, but incomplete

For this section, we examined whether the EU action plan had been relevant when first drawn up, i.e. whether it addressed the needs identified by experts and other stakeholders. We also assessed whether it had been reviewed and updated. We looked at the events and sources it was based on, and assessed whether it contains adequate coordination arrangements for communication and the elements necessary for measuring implementation performance and ensuring accountability.

The EU action plan was broadly consistent with experts’ and stakeholders’ views on disinformation

Fighting disinformation is a highly technical domain that requires input and expertise from a diverse range of professionals. Public consultation is also essential to establish the views and priorities of stakeholders and to understand the threat better.

We found that the Commission had relied on appropriate external expertise and undertaken a comprehensive public consultation¹³ as a basis for the EU action plan. The EU action plan largely addressed the suggestions and concerns expressed in these documents.

At the time the EU action plan was published in December 2018, it presented a structured approach to address issues requiring both reactive (debunking and reducing the visibility of disinformation content) and proactive longer-term efforts (media literacy and measures to improve societal resilience). It emphasised the goal of protecting the upcoming 2019 European elections, as well as long-term societal challenges requiring the involvement of many different actors.

Except in the area of media literacy, the Commission identified concrete actions to follow the main recommendations from the report of the independent ‘High-level expert group on fake news and online disinformation’ (HLEG). The HLEG, consisting of 39 experts with different backgrounds, was set up in January 2018 by the Commission to advise on policy initiatives to counter disinformation online. This report, together with the Commission’s April 2018 communication formed the basis for the EU action plan.

Further evidence of the EU action plan’s relevance is that its actions sought to engage a broad array of key stakeholders in the area, including not only EU institutions and Member States, but also others such as the private sector, civil society, fact-checkers, journalists and academia.

The EEAS and the Commission did not establish clear coordination arrangements to implement the EU action plan

The EU action plan against disinformation was not accompanied by an overall coordination framework to ensure that any EU response is effective and proportionate to the type and scale of the threat. Defining and coordinating communication workflows, for example, would make it possible to identify when to work together and in partnership with local actors and civil society to raise awareness about disinformation threats.

A communication strategy ensures a coherent response when different actors are involved. Each of the EU action plan’s four pillars is under the responsibility of a different Commission DG or the EEAS. This creates the risk of them ‘working in silos’ (i.e. in parallel without cooperation or coordination) when it comes to communication, without any single body being in charge or having complete oversight of all communication to combat disinformation.

The Commission’s Directorate-General for Communication (DG COMM) is responsible for external communication for the institution. DG COMM’s management plan for 2019 recognises its role in tackling disinformation and highlights the need for cooperation among DGs and other institutions, referring in particular to the Directorate-General for Communications Networks, Content and Technology (DG CNECT) and the Joint Research Centre. However, there is no mention of the EEAS or the StratCom task forces, which are also very much involved in positive communication and fighting disinformation.

DG COMM has established an internal network against disinformation (IND). Its aims include improving coordination of communication activities on tackling disinformation, establishing an online repository of validated rebuttals, systematically detecting disinformation and coordinating the response, and promoting positive messaging. Eleven meetings took place between its inception in May 2018 and January 2020. These meetings are broadly inclusive, with representatives from many Commission services and representations, the EEAS and other institutions, as well as other experts. However, the meetings so far have only involved representatives sharing information on action taken, with no connection to policy-making and no evidence of concrete follow-up actions or decisions taken to make the IND an effective coordination mechanism.

DG COMM’s 2019 management plan had only one indicator related to disinformation (out of 102) and this only measured the number of IND meetings.

In addition, the Commission representations play a vital role in the Commission’s external communication, through positive messaging and outreach, media briefings, debunking myths and fighting disinformation, and are expected to participate actively in the IND. Their myth-busting activities are listed on a page on each official representation website. These pages were often difficult to access as their location varied between representations: some (e.g. in Greece and Spain) included it under the news section, while others (e.g. in Poland and Ireland) did not. They were also not regularly updated. Some pages provided limited, often-anecdotal information and there were no statistics available on the number of visitors to these pages.

Finally, DG COMM was developing a central web-based disinformation hub as a centralised portal bringing together all aspects of the EU institutions’ work on disinformation. Its launch was due in early 2020 but it has been cancelled for unclear reasons.

A piecemeal monitoring and reporting framework and the lack of long-term funding undermine the EU action plan’s accountability

To ensure accountability, an action plan needs clear objectives and time-bound actions, accompanied by a number of indicators for monitoring performance. Provisions for financing, regular reporting, evaluation and revisions are also essential elements of an action plan.

Some objectives in the EU action plan have generic wording such as “step up” or “strengthen”, which do not lend themselves to measurement. No overall KPIs exist for the EU action plan as a whole. In addition, half of the actions (actions 1, 2, 4, 5 and 8) have no KPIs and the actions are either not clearly defined or time bound (see also Annex IV).

The timeframe of the actions varies between short- and long-term, and some are concrete and time-bound (e.g. “by March 2019, the Commission and the High Representative, in cooperation with Member States, will establish a Rapid Alert System”), while others are vague (e.g. “Member States should significantly strengthen their own communication efforts on Union values and policies”).

The EU action plan was not accompanied by a dedicated monitoring and evaluation framework (this applies also to the recently published European democracy action plan). There was no provision to evaluate the plan as a whole and there has been no overall evaluation to date. Feedback from its implementation in the Member States is not recorded centrally and not aggregated. Each representation carries out its own communication campaign and collects statistics, but we did not find any evidence of those statistics being used by the Commission for lessons learned, best practice, or as a baseline. There is no reporting beyond stating that some activities belong to the category of efforts against disinformation. For example, tools such as the Commission’s internal myth-busting wiki or the newsletter against disinformation are not monitored in terms of Member State engagement (e.g. using surveys, user statistics or indicators).

The Commission and EEAS regularly update and report to various Council working parties and preparatory bodies on progress in implementing actions under specific pillars of the EU action plan. However, this reporting is not in the public domain and does not encompass the whole EU action plan.

Although there have been separate reports on specific aspects of the EU action plan (an assessment of the code of practice and a Commission report on the EU elections), only one report on the implementation of the EU action plan as a whole has been published. This was on 14 June 2019, six months after the presentation of the EU action plan itself.

Although this first implementation report covered all pillars of the EU action plan, it has a number of shortcomings:

• it does not provide any measure of performance;
• except for the code of practice, reporting for each pillar is mostly in a general narrative form and there is no detailed reporting for each action;
• there is no reporting annex on individual projects linked to the EU action plan;
• there is no indication of when to expect the next implementation report.

Fighting disinformation is a constantly evolving area, which would merit regular reporting. The Joint Communication on tackling COVID‑19 disinformation recognised the need to develop regular reporting¹⁴.

The EU action plan lacks a dedicated financial plan covering the costs of all activities assigned to different entities. Financing comes from different sources and there are no provisions to secure funding in the long term, even though some of the events mentioned in the EU action plan are recurring. Annex II presents the budget allocated to the different actions to combat disinformation. It shows that the main source of funding is different every year and exposes a lack of financial planning (see also paragraphs 50-51). The Commission and the EEAS do not always earmark expenditure linked to fighting disinformation (it has no specific intervention code) – such information has been compiled only for this audit. Figure 2 gives an overview of all EU funding against disinformation from 2015 to 2020 (it does not include activities that contribute indirectly to combatting disinformation, namely the pro-active communications activities in the EU’s neighbourhood).

Source: ECA, based on Commission and EEAS information.

Furthermore, the EU action plan has not yet been updated since it was presented in 2018. For example, some actions are linked only to the 2019 European elections or the 2019 Media Literacy Week, both of which have passed. Disinformation is constantly evolving. The tactics used, the technology behind disinformation campaigns, and the actors involved are all constantly changing¹⁵. The Council also highlighted the need to regularly review and update the EU action plan (see paragraph 10).

Although the COVID‑19 disinformation communication (June 2020), the European democracy action plan and the proposal for a digital services act extend certain actions originally set out in the EU action plan, they cannot be considered a comprehensive update thereof. Moreover, having actions pursuing similar objectives in different action plans and initiatives makes coordination more complex, increasing the risk of inefficiencies.

The implementation of the EU action plan is broadly on track, but has revealed a number of shortcomings

This section assesses the implementation of the actions under each of the four pillars of the EU action plan and the extent to which they have improved the way the EU tackles disinformation.

The strategic communications task forces play an important role but are not adequately staffed or funded to deal with emerging threats

Under pillar I of the EU action plan, we examined the EEAS StratCom task forces. We looked at their mandate and determined whether they were adequately staffed and sufficiently funded. In this context, we also examined the role and position of EUvsDisinfo, an EU flagship project against disinformation.

The StratCom task forces’ mandates do not adequately cover the full range of disinformation actors

Apart from improving the EU’s capacity to forecast and respond to external disinformation activities (the StratCom task forces’ mandates do not cover disinformation generated within the EU), the StratCom task forces have contributed greatly to effective communication and promoting EU policies in neighbouring regions.

The mandates of the three StratCom task forces (East, Western Balkans and South) have grown out of a series of Council Conclusions, with differences in tasks and focus. For example, the mandate of the East StratCom task force specifically covers the task “challenge Russia’s ongoing disinformation campaigns”¹⁶. The East StratCom task force’s mandate was defined in terms of a single, external malign actor rather than protecting Europe from disinformation regardless of source.

This was not the case for the other two StratCom task forces, whose original focus was stepping up communication activities in their respective regions. The South StratCom task force was established to cover the EU’s southern neighbourhood and the Gulf region, while the Western Balkans task force was established to enhance strategic communications in that region. Prior to the December 2019 Council Conclusions¹⁷, addressing disinformation was not the central priority of either task force. Only the East StratCom task force had the explicit objective of strengthening the capacity to forecast, address and respond to disinformation. Table 2 below sets out each StratCom task force’s objectives at the time of the audit.

Source: EEAS.

The three StratCom task forces are distributed widely across different regions and cover different agents of disinformation. Nevertheless, the StratCom task forces’ media monitoring activities focus extensively on Russian international media, Russian official communication channels, proxy media, and/or media inspired/driven by the Russian narrative, operating in the EU and its neighbourhood. However, according to EEAS analysis, other actors such as China have emerged to varying degrees as prominent disinformation threats. In this respect, the new European Parliament committee on foreign interference (INGE) has also held hearings to discuss potential threats from third countries¹⁸.

The StratCom task forces’ mandate is a political one, which does not specifically spell out their policy objectives and is not underpinned by a firm legal foundation. Under action 2 of the EU action plan, the High Representative was meant to review the mandates of the South and West Balkans StratCom task forces (but not the East task force). However, this review was never done. The EEAS considers that the Council Conclusions adopted in December 2019, by stating explicitly that “all three Task Forces should be able to continuously detect, analyse and challenge disinformation activities”¹⁹, provides sufficient basis for (re-)affirming their mandates. The Council also invited the EEAS to assess its needs and options for expanding to other geographical areas, which demonstrates that political support exists to broaden the scope of the StratCom task forces.

The StratCom task forces do not have a dedicated and stable funding source

When it was set up in 2015, the East StratCom task force was not provided with any resources of its own and was funded out of the administrative expenditure of the EEAS and the Commission’s Service for Foreign Policy Instruments. The EU action plan boosted the funding available for the EEAS StratCom task forces. In fact, strategic communications is the only part of the action plan whose specific budget has increased. As Figure 2 above shows, the budget for the StratCom task forces and strategic communications has increased nearly fourfold since the action plan was adopted.

Even though disinformation is not just a short-term threat, the StratCom task forces do not have a stable funding source, which could threaten their sustainability. For example, a significant source of funding for the StratCom task forces has been a European Parliament preparatory action called ‘StratCom Plus’ (see Annex II). By their nature, preparatory actions are designed to prepare new actions like policies, legislation and programmes. Figure 3 illustrates how the added funding has been allocated to improving different capabilities.

Source: ECA, based on EEAS data.

The importance of funding and adequate resourcing has been stressed on numerous occasions²⁰, including by the European Parliament²¹, Member States²² and civil society²³. However, stakeholders differ in their opinions about how to prioritise the available EU funding to fight disinformation. Based on our interviews, some Member States feel greater emphasis should be placed on analysing and monitoring those sources and actors to which disinformation can be more easily attributed. Others feel more funding should be allocated to positive communication.

Staffing needs not yet satisfied

The EU action plan envisaged strengthening the StratCom division by adding 11 positions ahead of the European elections, recruiting permanent officials in the medium term and new staff in the EU delegations “to reach a total increase of 50-55 staff members” by the end of 2020. The recruitment plan has been implemented in three phases: (1) redeployment of contract staff within the EEAS; (2) recruitment of staff to the StratCom team; and (3) addition of staff across 27 EU delegations in the EU neighbourhood.

The StratCom division is still in the process of recruiting and deploying staff. As of October 2020, it had 37 staff and had therefore not yet reached the total increase of 50-55 staff, as stated in the EU action plan. One reason why it has been difficult to meet this target is that many StratCom staff are seconded from the Council, the Commission and the Member States and some have been withdrawn from their secondments.

Almost all staff reinforcements (including all the redeployments) have been contract staff: the EEAS acknowledged that it is not easy to recruit permanent officials with the necessary expertise and skills needed to perform the duties required. However, despite their important contribution, contract staff have a maximum tenure of six years.

The other group that makes up a significant portion of StratCom capacity is seconded national experts. These have primarily supported the work of the East and recently also the Western Balkans StratCom task force. As well as benefitting the EEAS, their secondment benefits their home countries by allowing them to obtain more expertise and create deeper connections with the EEAS. However, overreliance on secondment can result in uncertainty in staffing and periodic loss of institutional memory and expertise due to frequent turnover. All of these factors potentially undermine the building of institutional memory and expertise.

In light of the COVID‑19 pandemic and the additional workload it has created for the task forces, there is a risk that the EEAS, with the current distribution and number of staff, will have insufficient capacity to keep pace with new trends and developments like emerging threat sources and disinformation strategies and tactics. Moreover, the Council’s request to reinforce strategic communications work in other regions (see paragraph 10) may further stretch its limited staffing capacity.

Effective data analysis is critical not only for monitoring, detecting and analysing disinformation, but also as a basis for sound, evidence-based strategic insights and policy-making. At the time of the audit, the StratCom division’s data analysis cell includes full-time in-house analysts, who are supported by external contractors. The cell, launched in mid-2019, supports the work of the StratCom task forces and the rapid alert system (RAS) under pillar II of the EU action plan. Its analyses are predominantly on demand and ad hoc. In addition, its work is not integrated in a structured manner into the work of all the StratCom task forces. Although relying on external contractors can provide flexibility, in-house capacity is critical to provide sensitive analysis at short notice and to build institutional memory and expertise.

Measuring the impact of the StratCom task forces’ work remains a challenge

The greatest challenges in strategic communications remain measuring vulnerability to and the impact of disinformation, as well as efforts to understand, analyse and respond to it. The Commission uses opinion polling as one way to assess the effectiveness of strategic communications to influence perceptions about the EU. However, it is difficult to attribute the results of this polling to EU actions.

Beyond the communication campaigns, the StratCom task forces did not comprehensively measure the impact of their work. Furthermore, none of them had an evaluation function to assess their effectiveness and identify areas for improvement.

The hosting of EUvsDisinfo by the EEAS creates uncertainty about the project’s ultimate purpose

EUvsDisinfo is the public face and flagship of the EU’s efforts to combat disinformation, and the East task force’s main anti-disinformation product. It has a searchable, open-source database, with more than 9 700 disinformation cases as of 1 October 2020. Key materials on the EUvsDisinfo website are published in five EU languages; the rest is in English and Russian only. According to the EEAS, the original focus on Russian disinformation provided the foundation for a unique and pioneering approach; there are no comparable initiatives by Member State governments.

From its beginnings in 2015, EUvsDisinfo has steadily built up visibility online (see Figure 4 below) by cataloguing, analysing and publishing examples of Russian disinformation. Many stakeholders confirmed that EUvsDisinfo has been instrumental in raising awareness and influencing perceptions about the Russian disinformation threat to the EU and its Member States.

Source: ECA, based on EEAS information.

However, EUvsDisinfo has also faced criticism in the past. For example, it was censured by the Dutch parliament²⁴ in 2018 for erroneously attributing Russian disinformation to a Dutch domestic publication. Furthermore, some cases published on EUvsDisinfo do not represent a threat to EU democracies.

Looking ahead, the future role and mission of EUvsDisinfo are unclear, beyond producing more examples of Russian disinformation – a threat that is now well established and acknowledged. Despite the EEAS’s claims that EUvsDisinfo is independent and does not represent the EU’s official position, its location within the EEAS makes this questionable. This raises the question of whether such an instrument should be hosted and run by a public authority (such as the EEAS) or whether it should be run under the responsibility of a civil society organisation.

The RAS has brought Member States together but has not lived up to its full potential

The establishment of the rapid alert system (RAS) is the key action under pillar II of the EU action plan (action 3). As the plan states, “The first hours after disinformation is released, are critical for detecting, analysing and responding to it”. The RAS was established in March 2019, by the deadline set in the plan. It has two key elements: a network of national contact points and a web-based platform whose aim is to “to provide alerts on disinformation campaigns in real-time through a dedicated technological infrastructure” in order to “facilitate sharing of data and assessment, to enable common situational awareness, coordinated attribution and response and ensure time and resource efficiency”²⁵. The EEAS provides the secretariat for the RAS and hosts the website.

Although a useful tool for sharing information, the RAS had not yet issued alerts at the time of the audit and has not been used to coordinate common action

In order to work effectively, the RAS must be able to issue timely alerts, coordinate common attribution and response and facilitate exchange of information between Member States and EU institutions. We examined whether the RAS had been operational before the 2019 European elections, as stipulated in the EU action plan. We also assessed its activity and the level of engagement of its participants.

We found that the RAS was set up quickly in March 2019 as envisaged in the EU action plan. It has brought together Member States and EU institutions, and has facilitated information sharing, but it had not issued alerts at the time of the audit and has not coordinated common attribution and response as initially envisaged.

Most stakeholders consulted during our audit had a positive view of the RAS. For them, it fills an important gap in the counter-disinformation ecosystem by creating a community. This was also confirmed by our survey of Member States: the RAS allows them to share information, obtain new insights and mutually strengthen their capabilities. Figure 5 below presents the aspects most valued by the national RAS contact points.

Source: ECA.

Despite this positive opinion of the RAS as an information exchange tool, we found no evidence that the information shared through the RAS had triggered any substantial policy developments at Member State level. Building common situational awareness remains a work in progress for the RAS, hampered by the absence of harmonised and consistent definitions (e.g. of the term disinformation itself, and differing views on its sources, responses, levels of preparedness, etc.) and lack of common risk assessment.

When the RAS was launched, providing real-time alerts to react swiftly to disinformation campaigns was considered its primary purpose, driven by the urgency of the upcoming European elections. For the StratCom team, however, the key aim was to bring practitioners together and develop a community, as no such mechanism previously existed in the EU. These differing motives have obscured understanding among stakeholders and the wider public of what the RAS does.

An alert mechanism has been developed, which can be used in extremely urgent cases, but it had not been activated at the time of the audit. A threshold for triggering the alert system has been defined in qualitative terms: a disinformation campaign that has ‘transnational significant impact’ (i.e. a targeted attack on several countries). However, a quantitative assessment of this threshold is not possible.

In addition to the alert function, the RAS was conceived to help attribute disinformation attacks to their source and promote a coordinated response. However, this coordination capacity of the RAS has not been tested.

Activity and engagement in the RAS are driven by a limited number of Member States

The RAS brings together the contact points of the Member States, the Intelligence and Situation Centre of the EEAS, the European Commission (especially DGs CNECT, JUST and COMM), the European Parliament and the General Secretariat of the Council. Representatives from NATO, the G7 Rapid Response Mechanism participate in the rapid alert system. Outside experts, including from civil society and online platforms, are also sometimes present during RAS meetings. Generally, meetings of the national contact points take place every quarter, but the level of engagement varies between different Member States. Most activity is driven by one third of the Member States, which participate more regularly and are more active in the meetings.

Latest statistics point to a downward trend in activity levels

The statistics generated by the platform point to a number of trends. Firstly, activity is driven by a small number of core users, with other users being much more passive. Secondly, since its launch, activity levels have peaked around two main events: the European elections and the early weeks following the general lockdown in mid-March 2020. However, in the case of the latter, these have since come down and stabilised at the end of August 2020 at around half the levels in May.

The user statistics point to a downward trend in activity levels. For example, average daily views – even in the specific parts of the RAS that focus on COVID‑19 disinformation – have declined, as shown in Figure 6. In addition, the number of actively engaged users has been in steady decline since the European elections in late May 2019. While these metrics do not tell the full story, they indicate clearly that the platform is not living up to its full potential.

Source: EEAS StratCom. The two drops in the user number is a result of the policy introduced in August to deactivate user accounts that have not been active for more than three months.

Cooperation with online platforms and existing networks is mostly informal

According to the EU action plan, online platforms should cooperate with the RAS contact points, in particular during election periods, to provide relevant and timely information. However, there is no protocol establishing cooperation between the RAS and the online platforms and as the StratCom team does not monitor the number of cases flagged, it is not possible to assess the RAS’s performance in this area.

The code of practice made online platforms adopt a stance against disinformation but stopped short of making them accountable

One of the main reasons why disinformation is so acute is the widespread use of the internet, combined with the advent of new technologies and the ever-increasing use of online platforms to access information. This greatly facilitates the creation, amplification and dissemination of false information. According to the Digital Economy and Society Index, in 2020, 85 % of EU citizens were internet users. Most platforms monetise their services through their handling of personal data (mainly based on advertising models). This created a fertile ground also for disinformation actors, allowing them to better target their actions.

In the case of online platforms, disinformation occurs mostly as a result of users sharing false information, which can then be promoted by the platforms’ algorithms to prioritise the display of content. These algorithms are driven by the online platforms’ business model, and they privilege personalised and popular content, as it is more likely to attract attention. Disinformation also affects web search results, which further hinders users in finding and reading trustworthy online information²⁶. Picture 1 below shows an example of the search predictions provided by an online platform for the term “The EU is”, which are almost all negative.

Source: ECA actual internet search on 18 October 2019 at 11:55 (GMT +1).
Google is a trademark of Google LLC.

Fake accounts, internet trolls and malicious bots also contribute to the dissemination of false information.

The CoP provides a framework for the Commission to interact with social media platforms

Following the April 2018 Commission communication and the HLEG proposals, the Commission decided to engage with online platforms and other trade associations on the subject of disinformation. This led to the creation of the code of practice (CoP) (see Annex V), adopting a voluntary approach based on self-regulation by the signatories. The CoP was signed in October 2018, before being incorporated into the EU action plan under pillar III. Presently it has 16 signatories.

The CoP committed online platforms and trade associations representing the advertising sector to submit reports to the European Commission setting out the state of play regarding measures taken to comply with their commitments. These measures ranged from ensuring transparency in political advertising to closing fake accounts and preventing purveyors of disinformation from making money. The Commission closely monitored their compliance with these commitments.

Most of the stakeholders interviewed during the audit stressed that the Commission’s engagement with online platforms was a unique and necessary initiative. Many of those from outside the EU whom we consulted are observing the Commission’s efforts closely. They see the EU as the first global player trying to achieve the delicate balance between protecting freedom of expression and limiting malign spread of harmful disinformation.

The CoP provided a framework for the Commission to interact with social media platforms ahead of the EU elections in May 2019 and later during the COVID‑19 pandemic to mitigate the negative effects of the related “infodemic”. Box 1 presents EU efforts to limit COVID‑19 related disinformation through the CoP (see also Annex VI).

The assessment of the CoP revealed limitations in the reporting requirements

A number of reviews and evaluations have assessed the CoP. They revealed several shortcomings concerning the way the Commission established reporting requirements for the signatories to the CoP (see Box 2). These evaluations had not led to any changes in the CoP.

Our work confirms that signatories’ reporting varies, depending on their level of commitment and whether they are an online platform or a trade association. In addition, the online platforms’ reports are not always comparable and their length varies considerably.

This variation among signatories to the CoP also proved to be a problem for setting overall KPIs. These KPIs made it possible to monitor the actions of some signatories, but not all. For example, under the heading “Integrity of services”, the Commission proposed the indicator “Number of posts, images, videos or comments acted against for violation of platform policies on the misuse of automated bots”. This output indicator is relevant for specific online platforms only.

According to the Commission’s own analysis of the CoP signatories’ reports, the metrics provided so far satisfy only output indicators. For example, platforms report that they have rejected advertisements, or removed a number of accounts or messages that were vectors of disinformation in the context of COVID‑19 (see also Box 1). If this reported information is not put into context (i.e. by comparing it, in time, against baseline data and other relevant information such as the overall creation of accounts), and the Commission cannot verify its accuracy, it is of limited use.

The assessment of the CoP conducted on behalf of the Commission not only looks at the current state of the reporting but also recommends possible metrics for the future. The document proposes two levels of indicators:

1. ‘structural’ indicators for the code as a whole, measuring overall outcomes, the prevalence of disinformation online and the code’s impact in general. These help to monitor, at a general level, whether disinformation is on the rise, stable or declining;
2. tailored ‘service-level’ indicators, broken down by pillar, to measure each individual signatory platform’s results in combatting disinformation.

At the time of the audit, the Commission had not provided signatories with any new reporting template or more meaningful new indicators.

The issues described above show that online platforms are not held accountable for their actions and their role in actively tackling disinformation.

Lack of a coherent media literacy strategy and fragmentation of EU actions dilutes their impact

Pillar IV of the EU action plan focuses on raising awareness and strengthening societal resilience to disinformation. It aims to improve media literacy actions, such as the 2019 Media Literacy Week, and to support independent media and investigative journalists. It also calls on Member States to rapidly implement the media literacy provisions of the Audio-Visual Media Services Directive and create teams of multi-disciplinary independent fact-checkers in light of the 2019 European elections.

Media literacy’ refers to skills, knowledge and understanding that allow citizens to use media effectively and safely and equip them with the critical thinking skills needed to exercise judgment, analyse complex realities and distinguish between opinion and fact³³. Responsibility for media literacy, which lies at the cross-section of education policy and the EU’s digital agenda, lies with the Member States. The Commission’s role is to foster collaboration and facilitate progress in the area. Disinformation does not respect borders, however, and developing common tools and sharing best practices at EU level are important.

In order to assess the actions under this pillar, we examined the 2019 EU Media Literacy Week event, and whether there was a well-defined strategy for the various initiatives in this area. We reviewed the Commission’s report on the 2019 European elections³⁴ and assessed 20 projects directly related to media literacy and fighting disinformation.

The Commission’s report on the 2019 European elections stated that “[while] manipulative efforts recurrently focused on politically sensitive topics and targeted EU audiences ahead of the elections, no large-scale covert interference operation in the 2019 elections has been identified so far”.

Member States do not participate evenly in the EU Media Literacy Week

European Media Literacy Week is a series of actions to raise awareness about media literacy across the EU (see Box 3). It is unclear how it reflects a coherent EU media literacy strategy, however; although it includes some high-level discussions, it mainly serves to illustrate some specific EU and Member State initiatives. The 2020 edition was to be jointly organised between the Commission and the Council, which could have further stimulated Member State action and participation. However, it was cancelled due to COVID‑19.

Box 3

European Media Literacy Week

The 2019 Media Literacy Week was one of two specific media literacy actions in the EU action plan. It took place in March 2019 in Brussels and in Member States, and included a high-level conference. Over 320 events were organised at the time, and in total 360 events had been organised up to the end of September 2020.

Nearly half of all events took place in France, with Belgium (mostly Brussels) a distant second. A small number of Member States did not host any events at all – as the geographical distribution of the events shows (see picture opposite). As would be expected, most events took place around the time of the official launch. However, no further statistics are available on the number of people these events reached, their thematic distribution and the extent to which they dealt specifically with disinformation.

Source: ECA, based on Commission data.

There is no overarching media literacy strategy that includes tackling disinformation

We found there were a multitude of EU and Member State initiatives that addressed media literacy, and a plethora of policy documents. This is also clear from the Council Conclusions on media literacy³⁵, which include an annex with the main policy documents. However, these actions are not coordinated under an overarching strategy for strengthening societal resilience, particularly in media literacy, which would include tackling disinformation. While actions to address Member States’ specific challenges on media literacy are also important in order to achieve local impact, EU support for media literacy lacks the following underlying elements, which would be conducive to its sound financial management:

• a regular update , in cooperation with the Media Literacy Expert Group (MLEG), of the most important media literacy practices and actions in the EU and Member States (the Council of Europe produced such a mapping in 2016, the first of its kind; however, it has not been updated since then³⁶);
• clear objective-setting based on systematic and regular research into media literacy and the impact of media and digital platforms, accompanied by a set of indicators to measure performance;
• the necessary coordination mechanisms to create synergies and avoid overlap between initiatives and actions under, for example, the Audio-Visual Media Services Directive, the digital education action plan, the Creative Europe Programme, the digital competence framework and skills agenda, the recently published European democracy action plan and the Digital Services Act, the media and audiovisual action plan, etc.;
• unified monitoring of EU media literacy initiatives.

For the next multi annual financial framework (2021-2027), according to the Commission, approximately €14 million in EU funding from the Creative Europe programme³⁷ – €2 million per year – has been earmarked to support media literacy. However, as the Council Conclusions on media literacy also state, it will be necessary to develop additional funding sources.

Most projects examined produced tangible results, but many did not demonstrate sufficient scale and reach

Of the 20 projects we assessed, ten were funded under Horizon 2020, and the other ten were pilot projects and preparatory actions funded by the European Parliament (see table in Annex III).

The ‘Media Literacy for all’ call for proposals, launched in 2016 by the European Parliament, includes pilot projects and preparatory actions for co-financing innovative start-up ideas from across the EU in the field of media literacy. A pilot project runs for two years, followed by a three-year preparatory action. The Horizon 2020 projects are research and innovation projects covering highly technical aspects of tackling disinformation, such as the use and detection of bots or developing a new generation of content verification tools.

Our analysis (see Annex III) found tangible results in 12 out of 20 projects. Most positive results were achieved by projects that built on the results of previous projects to produce tools related to fact-checking, or by projects to create teaching and learning material against disinformation (see Box 4).

However, we identified shortcomings in 10 out of the 20 projects, which concerned mostly their small scale and limited reach. Seven projects did not or are unlikely to reach their intended audience, and the results achieved by three projects were difficult to replicate, which limited their impact. Box 5 presents a number of projects illustrating these issues:

Overall, there was little evidence of any comparative analysis of project results – especially in terms of what had worked and why. Nor was there much evidence of the Commission coordinating the exchange of best practice and media literacy material across the EU. An evaluation framework is also lacking. Such a framework is critical to the long-term development of societal resilience, as it ensures that lessons learned feed directly into future actions, policy and strategy. Obtaining evidence of the direct impact of media literacy measures is difficult, and efforts are still at an early stage of development³⁸. The Council Conclusions on media literacy also called for the development of systematic criteria and evaluation processes, and a uniform and comparative methodology for Member States’ reporting on the development of media literacy³⁹.

The SOMA and EDMO projects attracted limited interest from media literacy experts and fact-checkers

As the EU action plan states under pillar IV, independent fact-checkers and researchers play a key role in furthering the understanding of the structures that sustain disinformation and the mechanisms that shape how it is disseminated online. The Commission financed the Social Observatory for Disinformation and Social Media Analysis (SOMA) project, a digital platform with the aim of forming the basis for a European network of fact-checkers. SOMA is funded under Horizon 2020, with an overall budget of nearly €990 000. The project started in November 2018 and is scheduled to end on 30 April 2021.

Our analysis revealed that SOMA managed to attract only two fact-checkers recognised by the International Fact Checking Network⁴⁰. At the time of the audit (October 2020), SOMA had 48 registered members. Several of those we contacted admitted to never using the SOMA platform. While the technology behind SOMA has been evaluated positively, the project is still not widely used by the fact-checking community.

Well before the end of the SOMA project, and without waiting for an evaluation to collect and apply lessons learned, the Commission in June 2020 launched the first phase (worth €2.5 million, running until the end of 2022) of the European Digital Media Observatory (EDMO). It aims to strengthen societal resilience by bringing together fact-checkers, media literacy experts, and academic researchers to understand and analyse disinformation, in collaboration with media organisations, online platforms and media literacy practitioners.

SOMA and EDMO therefore have partially overlapping objectives, and most contractors are involved in both projects at the same time. The evaluators of SOMA suggested merging the two projects but no formal links between them have yet been established. There is also a risk of overlap in their financing, as both projects are based on and use the same technology and commercial products.

EDMO has been presented as a holistic solution to tackling many of the societal challenges around disinformation. However, being in its infancy, its visibility among stakeholders is still limited, according to its management. It is too early to judge the effectiveness of EDMO. Nevertheless, given the limited awareness of EDMO among stakeholders, its achievements may not match its overly ambitious goals. Its current focus is on building up the necessary infrastructure, and it will need more resources to succeed in its aims.

The media literacy experts we interviewed commented that the media literacy community was not feeling sufficiently engaged with EDMO. EDMO’s advisory board features a broad array of expertise from across academia and journalism, reflecting the EU action plan’s key emphasis on strengthening fact-checking and supporting journalism. However, the media literacy community or civil society, who could provide useful links between academia and policy-making are under represented (two out of 19 experts).

Conclusions and recommendations

We examined whether the EU action plan against disinformation was relevant when drawn up and delivering its intended results. We conclude that the EU action plan was relevant but incomplete, and even though its implementation is broadly on track and there is evidence of positive developments, some results have not been delivered as intended.

We found that the EU action plan was consistent with experts’ and stakeholders’ views and priorities. It contains relevant, proactive and reactive measures to fight disinformation. However, even though disinformation tactics, actors and technology are constantly evolving, the EU action plan has not been updated since it was presented in 2018. In December 2020, the Commission published the European democracy action plan, which includes actions against disinformation, without clarifying exactly how it relates to the EU action plan against disinformation (see paragraphs 20-24 and 41-42).

The EU action plan does not include coordination arrangements to ensure that EU responses to disinformation are coherent and proportionate to the type and scale of the threat. Each of the EU action plan’s pillars is under the responsibility of a different Commission Directorate-General or the European External Action Service, without any single body being in charge or having complete oversight of communication activities (see paragraphs 25-31).

There is no monitoring, evaluation and reporting framework accompanying the EU action plan and the European democracy action plan, which undermines accountability. In particular, the plans include generic objectives that cannot be measured, several actions that are not time-bound and no provisions for evaluation. Only one report on the implementation of the EU action plan has been published, with limited information on performance. Without a comprehensive, regular review and update, it is difficult to ensure that EU efforts in this area are effective and remain relevant. Furthermore, there was no comprehensive information on the funding sources and the estimated costs of the planned actions (see paragraphs 32-40).

The European Commission should improve the coordination and accountability framework for its actions against disinformation by incorporating:

1. clear coordination and communication arrangements between the relevant services implementing the EU actions against disinformation;
2. a dedicated monitoring and evaluation framework containing clear, measurable and time-bound actions, as well as indicators to measure performance and evaluation provisions;
3. regular reporting on the implementation of the actions, including any necessary updates;
4. a summary of the main funding sources and expenditure made in implementing the actions.

Timeframe: end of 2021 for recommendations a) and b) and mid 2023 for recommendations c) and d)

Under pillar I of the EU action plan the three EEAS strategic communication task forces have improved the EU’s capacity to forecast and respond to disinformation activities and have contributed substantially to effective communication and promoting EU policies in neighbouring countries. The task forces’ mandates do not adequately cover the full range of disinformation actors, including new emerging threats (see paragraphs 45-49).

Staffing of the task forces largely depends on the secondment of national experts, making it more difficult for the EEAS to manage and retain staff. The StratCom team has not yet met its recruitment targets, and the COVID‑19 crisis has created additional workload. Furthermore, the task forces have no evaluation function to assess their effectiveness and identify areas for improvement (see paragraphs 53-58 and 60).

The EEAS should:

1. bring emerging disinformation threats to the Council’s attention. It should then review and clarify the policy objectives to be achieved by strategic communications division and its task forces.
2. reach the recruitment targets set in the EU action plan;
3. focus its human resources on the most sensitive tasks, such as threat analysis and threat evolution, and outsource less sensitive communication activities where these cannot be done in-house due to staff shortages;
4. undertake regular evaluations of the task forces’ operational activities, beyond their communication campaigns.

Timeframe: mid 2022

EUvsDisinfo has been instrumental in raising awareness about Russian disinformation. However, its placement inside the EEAS raises some questions about its independence and ultimate purpose, as it could be perceived as representing the EU’s official position (see paragraphs 61-64).

Under pillar II, the EEAS quickly set up the rapid alert system (RAS). We found that the RAS had facilitated information sharing among Member States and EU institutions. However, the RAS has never issued alerts and, consequently, has not been used to coordinate joint attribution and response as initially envisaged. Furthermore, the latest statistics show that activity and engagement in the RAS are driven by a limited number of Member States. There is a downward trend in activity levels and cooperation with online platforms and existing networks is mostly informal. Additionally, there is no protocol establishing cooperation between the RAS and the online platforms (see paragraphs 65-76).

The EEAS should:

1. request detailed feedback from Member States on the reasons for their low level of engagement and take the necessary operational steps to address them;
2. use the RAS as a system for joint responses to disinformation and for coordinated action, as initially intended;
3. propose to the online platforms and the Member States a framework for cooperation between the RAS and the online platforms.

Timeframe: mid 2022

The one action under pillar III is about ensuring continuous monitoring of the code of practice (CoP). It sets out a number of voluntary measures to be taken by online platforms and trade associations representing the advertising sector. With the CoP, the Commission has created a pioneering framework for engagement with online platforms. During the initial stages of the COVID‑19 pandemic, the CoP led the platforms to give greater prominence to information from authoritative sources.

Our assessment of the CoP and the Commission’s evaluations revealed different reporting from the platforms, depending on their level of commitment. Moreover, the metrics that the platforms are required to report satisfy only output indicators. Platforms do not provide access to data sets, so that the Commission cannot verify the reported information. Consequently, the CoP falls short of its goal to hold online platforms accountable for their actions and their role in actively tackling disinformation (see paragraphs 77-89).

Building on recent initiatives such as the new European democracy action plan, the Commission should:

1. propose additional commitments to the signatories to address weaknesses identified in the evaluations of the code of practice;
2. improve the monitoring of the online platforms’ activities to tackle disinformation by setting meaningful KPIs;
3. establish a procedure for validating the information provided by online platforms.

Timeframe: end of 2021

Under pillar IV of the EU action plan, we found there were a multitude of EU and Member State media literacy initiatives, and a plethora of policy documents, which are not organised under an overall media literacy strategy that includes tackling disinformation (see paragraph 95).

Most activities during the 2019 EU Media Literacy Week took place in just two Member States, which substantially limited the initiative’s awareness-raising potential. Our analysis of a sample of 20 projects addressing disinformation found tangible results in 12 projects. Most positive results were achieved by projects that built on the results of previous projects to produce fact-checking tools or teaching material. The key shortcomings identified in 10 projects relate to the small scale and reach of their activities (see Box 3, paragraphs 94 and 97-101).

The Commission should adopt a media literacy strategy that includes combatting disinformation as an integral part. In order to better target disinformation through media literacy actions and reduce their fragmentation, this strategy should include:

1. a regular update, in cooperation with the Media Literacy Expert Group (MLEG), of the most important media literacy practices and actions in the EU and Member States;
2. clear objective setting based on systematic and regular research into media literacy and the impact of media and digital platforms, accompanied by a set of indicators to measure performance;
3. the necessary coordination mechanisms to create synergies among projects.

Timeframe: end of 2022

Independent fact-checkers and researchers play a key role in furthering the understanding of disinformation. So far, the Commission’s efforts to develop a European network of fact-checkers (the Social Observatory for disinformation and Social Media Analysis (SOMA)) have not managed to attract much awareness from this community. The European Digital Media Observatory (EDMO) is supposed to replace SOMA, but both projects were running in parallel at the time of the audit. EDMO’s visibility among stakeholders is still limited, which contrasts with its ambition of providing a holistic solution to the societal challenges around disinformation. Furthermore, the media literacy community and civil society, who could provide important links between academia and policy-making, are not well represented on EDMO’s advisory board (see paragraphs 102-107).

To ensure that EDMO meets its ambitious objectives, the Commission should:

1. collect lessons learned from the SOMA project once it has ended and apply them in EDMO;
2. increase the representation of media literacy and civil society experts in EDMO’s advisory board;
3. increase awareness of EDMO among stakeholders, in particular fact-checkers and media literacy experts;

Timeframe: end of 2021

This Report was adopted by Chamber III, headed by Mrs Bettina Jakobsen, Member of the Court of Auditors, in Luxembourg on 27 April 2021.

For the Court of Auditors

Klaus-Heiner Lehne
President

Annexes

Annex I — Main EU institution departments and offices fighting disinformation

Source: European Commission.

Annex II — EU spending on actions against disinformation (in euros)

Source: ECA on information provided by the Commission and the EEAS.

Annex III — Assessment of projects against disinformation (Pilot Projects, Preparatory Actions, H2020)

Source: ECA.

Annex IV — Assessment of the actions included in the EU action plan on Disinformation

Source: ECA.

Annex V — The code of practice on disinformation

The code of practice on disinformation (CoP) is a set of commitments to fight disinformation, agreed on a voluntary basis by representatives of online platforms, leading social networks, advertisers and the advertising industry. The CoP was encouraged and facilitated by the European Commission. This was the first time major industry players had decided to act together to address the spread of online disinformation.

The initial signatories in October 2018 included Facebook, Google, Twitter and Mozilla, as well as a number of advertisers and advertising industry bodies. Microsoft signed the CoP in May 2019, followed by TikTok in June 2020. Presently there are 16 signatories.

The CoP consists of a series of commitments under five pillars:

• scrutiny of ad placements;
• political advertising and issue-based advertising;
• integrity of services;
• empowering consumers;
• empowering the research community.

It also includes an annex identifying best practices that signatories will apply to implement the CoP’s commitments. The signatories prepared individual roadmaps to implement the CoP.

As the CoP is voluntary, there are no penalties for failure to honour these commitments. Therefore, it is important to monitor the signatories’ progress in implementing their commitments. Between January and May 2019, the European Commission verified the implementation of these commitments by Facebook, Google and Twitter, particularly in relation to the integrity of the European elections. The three platforms reported on a monthly basis on action they had taken in relation to the scrutiny of ad placements, transparency of political and issue-based advertising, and fake accounts and malicious use of bots. The Commission also invoked the CoP to request five sets of reports from the signatories on action they had taken to counter disinformation during the COVID‑19 pandemic.

These reports are available on the Commission’s website, together with the Commission’s assessment.

Annex VI — Chronology of the main EU actions as a response to the 2020 COVID-19 pandemic and “infodemic”

Source: ECA.

Acronyms and abbreviations

CoP: Code of Practice

EDMO: European Digital Media Observatory

EEAS: European External Action Service

EMLW: European Media Literacy Week

ERGA: European Regulators Group for Audio-visual Media Services

EU action plan: EU action plan against disinformation

EUD: EU Delegations

EUvsDisinfo: The flagship project of the European External Action Service’s East StratCom Task Force. It was established in 2015 to better forecast, address, and respond to the Russian Federation’s ongoing disinformation campaigns affecting the EU, its Member States, and countries in the shared neighbourhood.

HLEG: Independent High Level Expert Group on Fake News and Online Disinformation

HWP: Horizontal Working Party on Enhancing Resilience and Countering Hybrid Threats

IE: Information Environment

IND: Internal Network on Disinformation

RAS: Rapid Alert System

SOMA: Social Observatory for Disinformation and Social Media Analysis

StratCom: Strategic Communication

Glossary

Algorithms: A process or set of rules applied by a computer in calculations or other problem-solving operations.

Bot: An automated software application that is programmed to do certain tasks.

Disinformation: The communication of false or misleading information for the purpose of deceit.

Infodemic: An excessive amount of information – some accurate and some not – that can be used to obscure or distort the facts.

Internet troll: Person who posts insults, often laced with profanity or other offensive language on social networking sites.

Media literacy: Capacity to access, understand and interact with media and communications.

Misinformation: The communication of false or misleading information, whether in good faith or for the purpose of deceit.

Strategic Communications: Coordinated and coherent communication by an organisation in pursuit of specific goals.

Wiki: A collaborative website on which any user can add and edit content.

Audit team

The ECA’s special reports set out the results of its audits of EU policies and programmes, or of management-related topics from specific budgetary areas. The ECA selects and designs these audit tasks to be of maximum impact by considering the risks to performance or compliance, the level of income or spending involved, forthcoming developments and political and public interest.

This performance audit was carried out by Audit Chamber III External action, security and justice, headed by ECA Member Bettina Jakobsen. The audit was led by ECA Member Baudilio Tomé Muguruza, supported by Daniel Costa de Magalhaes, Head of Private Office and Ignacio Garcia de Parada, Private Office Attaché; Alejandro Ballester Gallardo, Principal Manager; Emmanuel-Douglas Hellinakis, Head of Task; Piotr Senator and Alexandre Tan, Auditors. Michael Pyper provided linguistic support.

Endnotes

¹ Communication on tackling online disinformation, COM(2018) 236 final of 26 April 2018.

² EUCO 11/15 (Point 13) European Council meeting (19 and 20 March 2015) – Conclusions.

³ Final report of the High Level Expert Group on Fake News and Online Disinformation.

⁴ See footnote 1.

⁵ EUCO 9/18 European Council meeting (28 June 2018) – Conclusions.

⁶ Articles 2-6 of Treaty on the functioning of the European Union.

⁷ Council Conclusions on the “Complementary efforts to enhance resilience and counter hybrid threats”, paragraph 30 (10 December 2019).

⁸ European Parliament resolution of 15 June 2017 on Online platforms and the Digital Single Market (2016/2276(INI)); European Parliament resolution of 3 May 2018 on media pluralism and media freedom in the European Union (2017/2209(INI); European Parliament resolution of 25 October 2018 on the use of Facebook users’ data by Cambridge Analytica and the impact on data protection (2018/2855(RSP)); European Parliament resolution on foreign electoral interference and disinformation in national and European democratic processes (2019/2810(RSP)).

⁹ World Health Organisation “Let’s flatten the infodemic curve”.

¹⁰ JOIN(2020) 8 final, 10 June 2020.

¹¹ Communication on the European Democracy Action Plan, COM(2020) 790 final of 3 December 2020.

¹² 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, COM(2020) 825 final of 15 December 2020.

¹³ European Commission public consultation on fake news and online disinformation.

¹⁴ Tackling COVID-19 disinformation – Getting the facts right, p. 7, JOIN(2020) 8 final of 10 June 2020.

¹⁵ “Automated tackling of disinformation”, EPRS study, March 2019.

¹⁶ European Council Conclusions on external relations (19 March 2015), Council document EUCO 11/15.

¹⁷ Council Conclusions on Complementary efforts to enhance resilience and counter hybrid threats (10 December 2019), Council document 14972/19.

¹⁸ European Parliament Hearing on foreign interference on 25 and 26 January 2021: China, Iran, India and Saudi Arabia and UAE; European Parliament Hearing on Foreign Interference on 1 February 2021: Turkey and Russia.

¹⁹ Complementary efforts to enhance resilience and counter hybrid threats – Council Conclusions (10 December 2019), Council document 14972/19.

²⁰ June 2018 European Council Conclusions and December 2019 Council Conclusions on complementary efforts to enhance resilience and counter hybrid threats.

²¹ P8_TA(2019) 0187, P9_TA(2019) 0031.

²² Information Manipulation: a Challenge for Our Democracies. A report by the Policy Planning Staff (CAPS, Ministry for Europe and Foreign Affairs) and the Institute for Strategic Research (IRSEM; Ministry of the Armed Forces) France, August 2018.

²³ For example, Democratic Defense against Disinformation, Atlantic Council, February 2018; GMF Policy Paper No.21, August 2019; Winning the Information War, CEPA, August 2016.

²⁴ Officiele bekendmakingen.

²⁵ RAS terms of reference.

²⁶ Automated Tackling of Disinformation, EPRS study, March 2019.

²⁷ First baseline reports – Fighting COVID-19 disinformation Monitoring Programme.

²⁸ Second set of reports – Fighting COVID-19 disinformation Monitoring Programme.

²⁹ Multi-stakeholder forum on disinformation.

³⁰ ERGA Report on disinformation: Assessment of the implementation of the Code of Practice.

³¹ Commission’s own evaluation of the code of practice.

³² Assessment of the Code of Practice on Disinformation – Achievements and areas for further improvement.

³³ Audiovisual Media Services Directive (AVMSD) (EU) 2018/1808.

³⁴ COM(2020) 252 final of 19 June 2020.

³⁵ See Annex to Council conclusions on media literacy in an ever-changing world (2020/C 193/06).

³⁶ Mapping of media literacy practices and actions in EU-28, European Audio-visual Observatory, Council of Europe, January 2016.

³⁷ Out of a total proposed budget of €61 million of the Cross-sectoral strand of the Creative Europe programme (budget line 07 05 03).

³⁸ Exploring Media Literacy Education as a Tool for Mitigating Truth Decay, RAND Corporation, January 2019.

³⁹ Council conclusions on media literacy in an ever-changing world (2020/C 193/06).

⁴⁰ Created by the Poynter Institute. Presently, it has 82 active members.

⁴¹ https://ec.europa.eu/info/sites/info/files/soteu2018-cybersecurity-elections-recommendation-5949_en.pdf; Reported in https://ec.europa.eu/info/files/com_2020_252_en.pdf_en

Timeline

Contact

EUROPEAN COURT OF AUDITORS
12, rue Alcide De Gasperi
1615 Luxembourg
LUXEMBOURG

Tel. +352 4398-1
Enquiries: eca.europa.eu/en/Pages/ContactForm.aspx
Website: eca.europa.eu
Twitter: @EUAuditors

More information on the European Union is available on the internet (http://europa.eu).

Luxembourg: Publications Office of the European Union, 2021

COPYRIGHT

© European Union, 2021.

The reuse policy of the European Court of Auditors (ECA) is implemented by Decision of the European Court of Auditors No 6-2019 on the open data policy and the reuse of documents.

Unless otherwise indicated (e.g. in individual copyright notices), the ECA’s content owned by the EU is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) licence. This means that reuse is allowed, provided appropriate credit is given and changes are indicated. The reuser must not distort the original meaning or message of the documents. The ECA shall not be liable for any consequences of reuse.

You are required to clear additional rights if a specific content depicts identifiable private individuals, e.g. in pictures of the ECA’s staff or includes third-party works. Where permission is obtained, such permission shall cancel and replace the above-mentioned general permission and shall clearly indicate any restrictions on use.

To use or reproduce content that is not owned by the EU, you may need to seek permission directly from the copyright holders.

Software or documents covered by industrial property rights, such as patents, trade marks, registered designs, logos and names, are excluded from the ECA’s reuse policy and are not licensed to you.

The European Union’s family of institutional Web Sites, within the europa.eu domain, provides links to third-party sites. Since the ECA has no control over them, you are encouraged to review their privacy and copyright policies.

Use of European Court of Auditors’ logo

The European Court of Auditors logo must not be used without the European Court of Auditors’ prior consent.

GETTING IN TOUCH WITH THE EU

In person
All over the European Union there are hundreds of Europe Direct Information Centres. You can find the address of the centre nearest you at: https://europa.eu/european-union/contact_en

On the phone or by e-mail
Europe Direct is a service that your questions about the European Union. You can contact this service

• by freephone: 00 800 6 7 8 9 10 11 (certain operators may charge for these calls),
• at the following standard number: +32 22999696 or
• by electronic mail via: https://europa.eu/european-union/contact_en

FINDING INFORMATION ABOUT THE EU

Online
Information about the European Union in all the official languages of the EU is available on the Europa website at: https://europa.eu/european-union/index_en

EU Publications
You can download or order free and priced EU publications at: https://op.europa.eu/en/web/general-publications/publications. Multiple copies of free publications may be obtained by contacting Europe Direct or your local information centre (see https://europa.eu/european-union/contact_en)

EU law and related documents
For access to legal information from the EU, including all EU law since 1952 in all the official language versions, go to EUR-Lex at: http://eur-lex.europa.eu/homepage.html?locale=en

Open data from the EU
The EU Open Data Portal (https://data.europa.eu/euodp/en/home) provides access to datasets from the EU. Data can be downloaded and reused for free, both for commercial and non-commercial purposes.