Monthly Archives: March 2019

On May 28-29 in Brussels, join the community working against disinformation: case studies, civil society initiatives and tools will be presented. Pre-registrations for EU Disinfolab conference are open. Take a look at the agenda.

Facebook attracts an angry audience

A new report of NewsWhip shows the effects of the new algorithms of Facebook. Findings reveal that “Angry” is the top reaction when it comes to politics content and generally, the engagement is much higher than in 2018. On the algorithm matter, Facebook algorithm personalisation has been discussed during DisinfoLab webinar with Facebook Tracking Exposed. A summary will soon be available on our website.

Authoritarian regimes ready for the New Age of Warfare

The man in charge of Saudi Arabia’s campaign to stifle dissent took actions to spy on people he was considering as threats to the kingdom. Today, in this new age of warfare, internet mercenaries battle for authoritarian government by seizing on the tools for darker purposes.In addition, a report by Reporters Without Borders investigates Beijing’s strategy to control information beyond its borders, which is being a threat for journalism and is challenging western liberal democracy more and more.

EU elections

Good, but can do better

After publishing the second monthly intermediate results of the EU Code of Practice against disinformation, the European Commission urges major social media platforms to take further measures. Although delivering good results, the platforms are now demanded to show key performance indicators and to ban fake accounts in order to prevent the spread of disinformation before the upcoming EU elections.

Moreover, the European Council has adopted its Conclusions on securing free and fair elections and fighting disinformation and urges online platforms and social networks to fully implement the Code of Practice and ensure higher standards of responsibility and transparency.

Spain is getting ready for the elections period

A team of 100 officers from the Spanish Ministry of Interior will scour social networks and the so-called deep-web from early April to fight intentionally misleading or wrongful information before, during or after national and European elections.

EU fact checks of the week

• New initiative FactCheckEU, gathering fact checkers from 19 EU Member States will monitor politicians’ rhetoric and misinformation ahead of the May parliamentary elections.
• Another fact checking initiative to keep in favorites: EUFactCheck, led by journalist students. They have recently proven false the statement of the Dutch right-wing leader Thierry Baudet, about the increased homophobia in the Netherlands caused by uncontrolled immigration and open borders.
• A fake news about free brothels coupons for migrants spread from Germany to Greece.

Library

• European Parliament’s Science and Technology Assessment Panel has published a report drafted by EU DisinfoLab on challenges around the automated tackling of disinformation. (Summary available here).
• A study commissioned by the LIBE Committee of the European Parliament shows how strategic political propaganda through social media can cause a threat to the rule of law in the EU and its Member States.
• Instagram’s recommendation engine has let to anti-vaccine accounts promotion.
• In-depth look at whether the so-called “backfire effect” really exists and what it means for fact checkers.  
• Bellingcat played “Catch me if you can” on Instagram with a fugitive convict from the Netherlands. Thanks to the help of 60 Twitter users, the location of the criminal was revealed.

Calendar and announcements

Congratulations to DW Innovation, Fraun Hofer, and iLab ATC: their project on audio deepfake detection has been awarded by DNI Fund of Google. Here you can find all the DNI-funded projects.

See all past and upcoming events in our agenda

HR Corner

• EU H2020 project EUNOMIA is seeking for a postdoctoral research fellow in data analytics and machine learning for information trustworthiness.
• The Department of Politics and International Relations of the University of Oxford is looking for a Postdoctoral Research Fellow in Digital News.
• Warsaw Euro-Atlantic Summer Academy have opened its application process for the summer program on the topic of “4Humans, Machines, Politics, Truth”.

Our weekly newsletter on disinformation issues.

You wish to receive this by email? Subscribe here.

On May 28-29 in Brussels, join the community working against disinformation: case studies, civil society initiatives and tools will be presented. Pre-registrations for EU Disinfolab conference are open. Take a look at the agenda.

EU elections: The empire strikes back

On 12th of March, the European Parliament has adopted the amended regulation on the protection of personal data in the context of elections to the European Parliament. The adoption of the final text by the Member States is planned for 19th of March. This text is part of a package for securing free and fair elections. Moreover, in another resolution (a non-binding decision), parliamentarians urge the EU to prepare itself to fight back hostile disinformation campaigns of Russia along with China, Iran and North Korea.

Moreover, the European Data Protection Board has adopted a Statement on the use of personal data in the course of political campaigns and it is to note that today, the 18th of March, the rapid alert system between EU Member States is entering into force.

Can the web be fixed?

In an interview for the 30 years of the internet, Sir Tim Berners-Lee qualifies manipulation of information online as “cybercrime”, that should be combated by a refoundation of the web, updated with fresh legal and technical rules. The inventor of the World Wide Web is urging governments to protect people’s rights and freedoms online by translating laws and regulations for the digital age.

Brexit and far-right supporters: deal or no deal?

Just when the British parliament rejected a no-deal Brexit last week, a study shows how bots and automated tweets have amplified specific messages from a very small number of users during Brexit. According to the study of F-Secure, these tweets mainly came from outside of the UK and from accounts promoting a far-right agenda.Another study of the University of Cambridge and YouGov shows that pro-leavers are more likely to believe at least one conspiracy theory about how the country is run. According to the findings, 47% of leave voters in the UK and 44% of Trump voters in the US believe the government is deliberately hiding the truth about the number of immigrants in their country.

Library

What to read, watch and listen to this week:  

• European Parliament has published a report drafted by EU DisinfoLab on challenges around the automated tackling of disinformation.
• Spain is getting ready for the upcoming elections by creating new special intelligence unit guard against possible cyber-attacks and disinformation campaigns. 
• Attacks in New Zealand and how to avoid spreading misinformation and fear mongering.
• Elizabeth Warren has accused Facebook for censoring her posts, while Mrs. Warren did not respect the terms and conditions of the platform.
• Whatsapp will help users find out if an image is real or has been photo-shopped and even learn the background of an unaltered photo that has been stripped of its context.
• Chinese propaganda is now on Reddit.
• The line is too thin between empirical and emotional truth. Read the story about the real reasons behind posting fake news.
• Podcast: Adam Reichardt, Maciek Makulski and Jakub Kalenský discussing the differences between disinformation and misinformation, what the West is doing to combat fake news – all ahead of the May EU parliament elections.

Calendar and announcements 

• March 20 – EU DisinfoLab Webinar with Facebook Tracking Exposed: first initiative showing how Facebook feed algorithm is personalised. Link to video conference can be found here.
• March 21-22 @ Brussels: during EU Council meeting, the heads of state or government will review progress in tackling disinformation and the need to protect the democratic integrity of the European and national elections across the EU.
• March 22 @ Brussels: 30 years of the Web – where do we go next? The event organised as part of the European Commission’s Next Generation Internet initiative.
• March 18-22: European Media Literacy Week. Events will take place across all European member states.

See all past and upcoming events in our agenda

HR Corner

• EU H2020 project EUNOMIA is seeking for a postdoctoral research follow in data analytics and machine learning for information trustworthiness.
• The Department of Politics and International Relations of University of Oxford is looking for a Postdoctoral Research Fellow in Digital News.
• Specialized in open source reporting, Christiaan Triebert joins the investigative group Bellingcat.

On May 28-29 in Brussels, join the community working against disinformation: case studies, civil society initiatives and tools will be presented. Pre-registrations for EU Disinfolab conference are open. Take a look at the agenda.

Health misinformation online and how to deal with it

The algorithms of Amazon are recommending misinformation conspiracy and anti-vaccine propaganda.  In consequence, people can make harmful purchases that could potentially put their health in danger. The issue of anti-vaccine propaganda is also a particular concern in the United States, where lively hearings were held by the Congress. As described by the New York Times, in Oregon, the hospital stay of a child found to have tetanus due to non-vaccination costs 800 000$. Also under pressure, Facebook, announced they will reduce the visibility of anti-vaccine propaganda.

Alt-right US billionaires funding misinformation campaigns in Europe

Since a few years, a small group of American billionaires, financing the right wing of the Republican Party, have supported campaigns to disseminate false information in several EU countries. Small groups of activists and specialized political communication companies were funded, along with the purchase of ads on social networks to amplify conservative messages (such as the anti-abortion campaign in Ireland).

Regulation from the Atlantic to the Urals

After the white paper release by the British DCMS inquiry committee of the House of Commons earlier this year, the House of Lords released last week its report: “Regulating in the digital world”.The document calls for creation of a new regulation authority, or the extension of Ofcom’s oversight powers to online platforms. In a broader way, the House of Lords considered regulators should be granted responsibility for enforcing a duty of care on those companies.

In the meantime, Russia’s parliament has passed two bills outlawing “disrespect” of authorities and the spreading of what the government deems to be “fake news”, the bill is supposed to be signed by the Russian president on March 13.

Disinfo week

Last week three European capitals: Athens, Madrid and Brussels hosted the Atlantic Council#DisinfoWeek. The events offered a series of strategic dialogues on how to address the challenge of disinformation. Experts and officials from both sides of the Atlantic such as Ben Nimmo, Alina Polyakova, Guillaume Chaslot and Damian Collins shared their analyse and strategies to tackle the phenomenon of disinformation, through AI, debunking storytelling and regulation. Videos of the event can be seen here.

Library

What to read this week:

• Mark Zuckerberg has promised to start working on an updated and more private version of messaging and data sharing on Facebook.
• Google’s decision to ban political ads did not please to the candidates of presidential elections in Canada
• US Ambassador to the EU promised to reinforce transatlantic cooperation on disinformation fight.
• Researchers of Cornell University have proven how the recommender systems of major media platforms can shape people’s mind, create filter bubbles, echo chambers and shift user’s world view.
• The elections in South Africa and how to track bots and fake news in political context.
• Facebook is clearing up inauthentic accounts on Facebook and Instagram platforms: DFR analysis of takedown of suspicious Pakistani owned accounts.
• According to Jochai Ben-Avie, senior global policy manager for Mozilla, India should look at Europe as its model for data privacy GDPR, instead of turning to Chinese-style rules to control citizen’s data.

Calendar and announcement

• March 20 – EU DisinfoLab Webinar: Collecting data to research algorithms personalization. Link to videoconference can be found here.

See all past and upcoming events in our agenda

HR Corner

• EU H2020 project EUNOMIA is seeking for a postdoctoral research follow in data analytics and machine learning for information trustworthiness.

More details, please

The European Commission has published reports by Facebook, Google and Twitter covering the progress made in January 2019 on their commitments to fight disinformation (most of them were expected, as explained in our previous newsletter of 25/02). Homework half-heartedly done because these platforms have not provided enough metrics to clearly measure the results of activities undertaken, especially with the scrutiny of ads placements. Next reports will be published in March 2019 and platforms already have a task assigned. They are expected to implement effective policies to ensure the integrity of electoral process for the European elections due in May 2019. 

Cybersecurity first

As part of the elections integrity toolkit, strong cybersecurity is key to ensuring the transparent functioning of the election system. In this context, the EU cybersecurity agency – ENISA- has released an opinion on the cybersecurity during electoral processes. Among its recommendation, the agency suggests that “Member States should consider introducing national legislation to tackle the challenges associated with online disinformation while protecting to the maximum extent possible the values set down in the Treaty of Lisbon and the Charter of Fundamental Rights of the EU”. 
Last week, Microsoft announced it detected attacks targeting think tanks working on democratic integrity in Europe. As part of its “defending democracy” programme, the company announced it will extend its “AccountGuard” programme in several EU countries in order to provide protection to political candidates, parties and also think tanks and non-profits working on issues related to democracy and electoral integrity.  

Moderation at all costs?

An article published in The Verge rose emotion over the working conditions of content moderators. Major tech platforms outsource content moderation to “process executives” working under difficult management and traumatizing mental conditions. From developing PTSD-like symptoms and even beginning to embrace the fringe viewpoints of the videos and memes, human moderation does have impactful consequences. This raises the issue of content moderation processes that should imply both AI and humans without ultimately being dehumanizing outcomes. A shared framework and guidelines between platforms and contractors could be a strong element of discussion.

Number of the week

77.8% of Europeans said they believed fake news and “politically motivated disinformation” posed a threat to the legitimacy of European elections according to a poll by Yougov for Avaaz. 

Library

What to read this week:

• New podcast Power 3.0 “The intersection of globalization and polarization: a conversation with Moisés Naím”: in this episode, featured guest Moisés Naím discusses how illiberal political actors and authoritarian regimes are leveraging the increasingly complex globalized information space to exploit societal cleavages created by political polarization, employing a problematic mix of media manipulation and disinformation.
• Google says it’s fighting misinformation, but how hard? Mathew Ingram investigates the efforts announced around the media of the company and point several weaknesses, especially regarding Youtube. 
• Facebook won’t renew two-thirds of existing Facebook Watch news shows: Facebook is showing a greater willingness to ax news programming that’s not working on Facebook Watch.
• ‘Deep fake’ videos threaten the world order: Nina Schick argues we must prepare for an age when AI allows anyone with a grudge to create convincing bogus clips.
• Beyond fixing Facebook: authors from Free Press make the case for a taxation supporting innovative journalism.
• Amazon anti-vax problem: how recommendations on Amazon can drive you to conspiracy theories.
• How to build a disinformation business: meet the man behind one of the Czech Republic’s most influential fake-media sites.

Calendar and announcements

• Pre-registrations for EU Disinfolab conference on May 28-29 in Brussels are open Pre-register
• March 4-8 the Atlantic Council is organising its Disinfoweek in Madrid, Brussels and Athens
• March 6 @ European Parliament – David Alandete will present his new book “Fake news: La nueva arma de destrucción masiva”
• Call for papers: Nordic network on the study of online disinformation
• EU project “EU algorithms” is seeking your input on emerging findings and knowledge gaps regarding the relationship between AI and disinfo. 

See all past and upcoming events in our agenda

Platforms progress

Implementation reports from signatories of the EU code of practice are expected to be published by the European Commission this week. Here’s what to expect:

Twitter is expanding transparency around political ads. An ad endorsing a party or a candidate will be viewable in Twitter Ads Transparency Centre, with details such as billing information, ad spend and impressions data per Tweet. Enforcement of this policy will begin on 11 March, after which only certified advertisers will be allowed to run political campaigning ads.

Google has released a white paper on the sidelines of Munich cybersecurity conference. This paper summarizes in a single document the efforts across the products of the company in fighting intentional spread of disinformation. Nothing much new, but a good recap.

Facebook: Also in Munich, Nathaniel Gleicher, Facebook head of cybersecurity, hold a discussion with DFR Lab Graham Brookie, discussing the importance of industry partnerships.

Zuckerberg fireside chat part II

As expected in his new year wishes, Mark Zuckerberg hosted its first discussion with thinkers on tech and society. In a conversation with Harvard law professor Jonathan Zittrain, Zuckerberg said that Facebook has been developing private communication tools since the service’s early days at Harvard, and that the company continues to focus on improving them.

Discussing the problem of disinformation on the platform at 0:38, Zuckerberg explains that it should not be up to Facebook to decide what is true or false, a reason why the company relies on the work of fact-checkers. According to him, the challenge is to strike a difficult balance between limiting the spread of misinformation and not positioning itself as arbiter of the truth.

How to cure the virus?

Anti-vaccine theories have been spreading online just as a virus. Information discouraging people from vaccinating their children might indeed have contributed to an increase in several diseases. In fact, the conspiracies on vaccine are getting more engagement on Facebook, despite being fact-checked.

In this context, politicians are trying to pressure social platforms to act and stop recommending such content. Also related to health misinformation, House Democrats are pressing Facebook on the privacy protection it has in place for people who share sensitive health information in groups.

As Google and Facebook struggle on the measures to take, Pinterest has stopped showing vaccination results as it was showing false health claims. Users can still pin fringe images to their own boards, but they can no longer use Pinterest for free viral distribution. This approach can be referred to as “freedom of speech versus freedom of reach.” Casey Newton calls it a perfect r

The reason why Mickey Mouse left Youtube

Youtube has been under growing scrutiny for its recommendation algorithms, which according to some might lead to clickbait videos and conspiracy theories. In the New York Times, Kevin Roose comes back to the booming of conspiracy theories on the platform. Walt Disney withdrew its advertising spending from YouTube, joining other companies including Nestle, after a blogger detailed how comments on Google’s video site were being used to facilitate a “soft-core pedophilia ring.” This move is particularly interesting as advertisers are increasingly scrutinized for their support in the ad economy financing misleading content. (On this issue, join our Webinar on Wednesday to discuss online advertising with Johnny Ryan from Brave).

This reaction happens just when the video service announced it is tweaking its algorithms in the US to “begin reducing recommendations of borderline content and content that could misinform users in harmful ways”.

Library

What to read, watch and listen to this week:

• Coda story has launched its new coverage topic “authoritarian tech”: global reporting on how governments abuse the power and potential of emerging technologies such as AI, facial recognition, digital tracking and social media and will explore how these technologies are strengthening autocratic regimes and stifling dissent.
• How to spot when news is fake? by Naja Bentzen for EP Think Tank.
• The story of activists who managed to manipulate Google to return the Pakistani flag when searching for the key-words “toilet paper” might have been a meme according to Google, who found no evidence of this query.
• An AI that writes convincing prose risks mass-producing fake news: the designers have opted not to release the full results of their research while they pause to consider the ethical implications of the technology.
• Do not trust even if you hear it: WhatsApp audios that are mobs about immigrants: in Spain, Whatsapp is used to share misleading informations about migrants.
• Zeynep Tufekci in Wired: the imperfect truth about finding facts in a world of fakes.
• ‘No image can be taken on face value’: Fake photos flood social media after a terrorist attack in India in Poynter.

Calendar and announcements

• 27 February – Disinfolab Webinar. The ad economy of disinformation: does an alternative model exist? With Johnny Ryan from Brave.
• Call for Nominations – 2019 GUE/NGL Award for Journalists, Whistleblowers and Defenders of the Right to Information: applications until 01/03.
• 19-20 March @ Dresden. Dresden Forum for International Politics 2019: “Participation or Polarisation? Social media and societal peace”.

The EU announced it will make 2.2 Million EUR available for three separate pilot projects in the area of media freedom and pluralism geared at promoting quality journalism and cross-border cooperation between media professionals and media councils in the digital age.

See all past and upcoming events in our agenda

HR Corner

Full Fact is looking for:

• A Designer to strengthen our visual identity
• A Product Manager to overhaul our website
• A Web Developer to refresh our website
• A Policy and Impact Officer to turn our fact checking into action

More open positions in our past newsletters