From Fake News to False Elections
When Gabon experienced an attempted coup d’etat in late 2018, very few media outlets picked up on the fact that suspicion of a deepfake fuelled underlying unrest. The fact that there was no foul play did not matter—the televised address of president Ali Bongo had appeared to be digitally altered and shots were fired. Deepfakes have sometimes been described as the Photoshop of video. While its creation is far more complex, it is a fairly accurate description of the end result.
Now, some may scoff at the idea that a deepfake could result in a coup d’etat in Europe. But not so fast. The influence a deepfake can have on the political process may be far more sinister. If we consider that deepfakes will be able to replace “traditional” fake news, we need to look at the role of the latter. Fake news has increasingly been used to exacerbate existing tensions within countries and stroke fear of immigrants. The onslaught of fake news, both within the EU, as well as within its direct sphere of influence, might therefore be classified as an attempt to undermine the European project. And while deepfakes are a mere tool in the toolbox of propaganda, their realistic appearance can make them extremely effective in achieving their goal.
But even when there is no malintent, deepfakes run the risk of influencing people unduly. Last September, Striscia la notizia, an Italian satirical news show, published a deepfake video of Matteo Renzi, the former Prime Minister, insulting other politicians and making vulgar gestures. While, for the most part, the video was recognized as satire due to its outrageous content, some social media users took offense, thinking it was real. The show has since released a “behind the scenes” clip of how a deepfake video is made, but continues to create deepfake videos, absent any widespread debate surrounding the matter. The show’s creators said they like to provoke and use their creative license, and Matteo Renzi himself took no offense. However, Corriere della Serra newspaper journalist Massimo Gramellini was among the few acknowledging the potential of deepfakes to influence politics:
Not only can deepfakes destroy someone’s reputation, but these sophisticated and instantaneous ways of manipulating reality reach millions of users who aren’t prepared for what they are seeing. Deepfakes end up making people both sceptical and gullible at the same time. We doubt everything yet also believe in something, often something entirely far-fetched.
Gramellini hints at another problem: there is a risk that unfavorable, yet real events, can be labelled as fake. While deepfakes may be brought into the world in order to start the rumor mill, it is now possible also to dismiss real footage as a deepfake in an attempt to deflect or discredit. Chesney and Citron call this the Liar’s Dividend. In essence, it is accusing real footage of being fake in order to escape the potential costs exposure would bring by casting doubt on what is real or not.
Theo Gevers, professor of Computer Vision at the University of Amsterdam, underscores this danger in Dutch newspaper Trouw “When you cannot distinguish real from fake content, you can no longer use the truth as a starting point,” he argues, while adding that the use of deepfakes goes beyond the innocent persiflage or twisted truth. We have already seen fake news making an appearance in the murder trial of Vitaly Markiv in Italy. Prosecutors included material published by Russia Today and Russkaya Vesna, both outlets that are generally considered to publish Russian propaganda as evidence. While these specific items in the evidence file were not considered to be decisive in the conviction of Markiv, it raises the question of what else might be relied upon as evidence. A deepfake perhaps? This is an especially worrying trend in a country which, according to its own Authority of Communication, is vulnerable to a growing amount of disinformation.
Deepfakes may become the next tool in the spread of fake news. The fact that attempts to derail the referendum campaign in North Macedonia with fake news (which urged people to vote against the governmental proposal) seems to have been partially successful, should be cause for concern for the legitimacy of elections in Europe. Similarly, Ukraine’s 2019 presidential campaign was also threatened by fake news campaigns discrediting candidates who were not pro-Russia. Having learned from the US presidential campaigns, the Ukrainian government as well as social media companies were more prepared than in the case of the US, which lead to the pre-emptive shutdown of a lot of fake social media accounts used to peddle fake news. Nevertheless, reports came out of Ukrainians being paid for others to use their Facebook accounts in an attempt to circumvent preventative measures, which drives home the point that those wishing to interfere in political campaigns will go through great lengths to achieve their goals. Deepfakes may just be the next step and can indeed exacerbate that problem.
Knowing about the existence of deepfakes should be our first line of defense, but how can we adequately safeguard our democratic processes from interference by this new tool for disseminating propaganda? Deepfakes are technologically such complex and fast evolving beings that we have not figured out how to tell the difference between a real and a well-made fake video. And every time we get close to uncovering tell-tale signs, such as limited blinking of the eyes, deepfakes get better, because they are created through a system that teaches itself. So perhaps the question should be: when it comes to preserving democracy, does it matter if we can tell a deepfake apart from a real video?
What does the law say?
Many wonder if using troll factories, bot armies, and deepfakes is legal. Can those wanting to influence the political playing field just buy that influence on social media? The answer to that question of course depends on the jurisdiction, but two points are important to emphasize here. First of all, in cyberspace it is notoriously difficult to adjudicate under which jurisdiction something might fall, given that a user may be in a different country than the server is, which may again be in a different country from where the company is registered as its main operating base.
Secondly, the law—almost without exception—lags behind the developments of the time. Harmful things are not always prohibited immediately and, in fact, the harm can be known for decades before legislation is put in place. Smoking cigarettes comes to mind as one example, but also the introduction of legislation to make wearing a seatbelt in cars mandatory. That is not always a bad thing. If deepfakes were to be immediately made illegal, it would overnight prohibit a host of legitimate applications, including those used in Hollywood to create the latest blockbusters.
Several jurisdictions have attempted to curb the proliferation of deepfakes. Both at the state, as well as at the federal level, the US is leading in the legislative battle against deepfakes. In Europe, however, it appears that there is little enthusiasm to legislate against deepfakes per se. It is unclear why that is the case, though it may have something to do with both the legal culture in the US, as well as the fact that many social media and online platforms on which deepfakes spread are headquartered in the US. That does not mean, however, that there are no legislative tools at the disposal of prosecutors to limit this type of abuse absent a specific deepfakes law. Germany, for example, has passed legislation obliging social media companies to remove harmful content within twenty-four hours of it being reported, which could also be used to limit the online proliferation of deepfakes. France, in an attempt to prevent deepfakes from interfering with elections, passed legislation in November of 2018, which provides judges with the power to order content amounting to disinformation to be immediately removed during election campaigns.However, this legislation only covers the three months leading up to national elections and it therefore did not apply to the municipal elections held in March 2020. So far, though, there is no indication that these elections were a target for orchestrated influence campaigns either. Furthermore, the decision in the only case brought before a French court so far stated that a gross exaggeration did not amount to disinformation as defined under the law.
Prohibiting content, however, is also a very slippery slope, since the right to freedom of speech and expression in Europe is firmly grounded in Article 10 of the European Convention on Human Rights, as well as internationally under Article 19 of the International Covenant on Civil and Political Rights (ICCPR). Important to note, however, is that freedom of speech or expression is not a right without its limitations. While the boundaries are sometimes interpreted differently, most countries legislate against hate speech and incitement to violence, have provisions which make slander punishable, and some prohibit speech which denies historical events, such as the Holocaust. Indeed, defamation, slander, and libel prohibitions, as well as copyright law, are also existing legal frameworks under which deepfakes may be considered a violation.
What do social media companies do?
Surely, though, if the law may not be able to prohibit fake user accounts spreading deepfakes intended to harm both people and political processes, the social media platforms through which they are spread should be able to intervene? That answer is complicated. Most social media platforms regulate against fake profiles, but are still slow in identifying them and shutting them down. When it comes to regulating content, this again gets into muddy waters with the right to freedom of expression, because the court of social media may come to a different conclusion than the court of law.
A more immediate tool would be to engage the social media networks in investigating how news goes viral. It has been suggested that there are differences between “natural” and “unnatural” viral news, which result in patterns that can be distinguished. Bots and troll factories do not function organically in spreading news as, for example, a meme that goes viral. Investing in research that can develop algorithms that may help in identifying “unnatural viral news” may help curb the spread of deepfakes before it has gone too far. The European Commission is commissioning studies in this field. However, access to the data the social media platforms hold is essential in order to study these phenomena, which requires significant buy-in from these platforms on which deepfakes can proliferate.
So, until such a buy-in can be secured, fact-checking remains the most prevalent tool in the toolbox of most social media companies. However, fact-checking and removing harmful or undesirable content is also outsourced to low wage countries. That means that those contractors may not be completely aware of some of the nuances of their work. While nudity may be a pretty straightforward thing to remove from, say, Facebook, hate speech may include regionally specific slang, which can be hard to identify without proper training.
Furthermore, fact-checking does not mean that social media platforms are not used, or abused, to sway popular opinion, as the UK referendum on Brexit illustrated. At the time of writing, the UK is yet to officially release the report of the official inquiry into disinformation campaigns and the influence they had over the 2016 Brexit vote. However, it is widely believed that the Russian Federation sponsored an extensive campaign stoking up fear of migrants and anti-EU sentiments in the lead up to the referendum vote in an attempt to sway the election outcome. This has led to attempts to launch public-private partnerships and incentives for corporations to curb the spread of disinformation online. For example, the European Commission, ahead of the 2019 European Parliament Elections, requested a Code of Practice be followed by social media platforms and the trade associations representing the advertising sector, aiming at increasing transparency surrounding (online) campaign ads. Furthermore, it “established an independent network of fact-checkers to increase the ability to detect and expose disinformation.”These initiatives were launched in addition to EU Member States’ own legal and social initiatives, such as the prohibition of micro-targeting and oversight mechanisms.
Can such a combination of legislation, voluntary commitments to Codes of Conduct, and overarching coordination help in stopping the spread of disinformation? Perhaps it is too soon to tell, but what is clear is that there are far too many loopholes for people to exploit. Facebook, for example, claimed that during the election campaign of the 2019 European Elections cross-border advertising was not possible; a claim that was proven wrong by the Dutch NGO Bits of Freedom. The NGO managed to purchase advertisements shown to Dutch voters while using a German Facebook account and a German bank account to settle the bill. This demonstrates that legislation passed in various EU countries that prohibit foreign entities to donate to political campaigns or sponsor political ads are not equipped to prevent such online circumvention of the law. While the report by Election Watch EU established thatdeepfake content had not been employed in this specific election cycle, it warned this could be the next level threat to democracy and called for more regulation and for social media platforms to provide researchers and oversight mechanisms with meaningful access to data.
The Liar’s Dividend – How do we respond?
Professor Gevers warns us that we are on the eve of something big and unless we take action at a national level, rather than sitting on our hands waiting for Facebook or “the Chinese” to come up with a defense against deep fakes, there will be dire consequences. Tom Van de Weghe is a Belgian journalist who spent a year at Stanford University studying deepfakes and their impact. Incidentally, the first political deepfake is accredited to Belgium; the political party SP.A created a Donald Trump deepfake in an attempt to get the discussion on climate change going.Van de Weghe believes that the proliferation of deepfakes is extremely dangerous, especially during election campaigns and goes so far as to call the money the US military is investing the detection of deepfakes part of an arms race. He adds that platforms like Facebook have an immense responsibility but argues for an approach which looks at the intention rather than the manipulation of video. He believes that measures should adequately consider satire, as well as digital manipulation that do not live up to the definition of a deepfake, such as the Nancy Pelosi cheap-fake.
But if technology is evolving so rapidly that it is hard to keep up with the developments, and the legislation cannot keep up, what can be done to stop this trend from influencing democratic elections? The answer is not to abandon the existing strategies, but to add to them. Perhaps the most effective manner of stopping deepfakes getting any traction is by teaching people digital awareness. While it may be a cumbersome project, one of which we will not reap the benefits for possibly a generation to come, it is the best line of defense in the face of a fast-evolving technology. Teaching children and young adults, as well as older, less tech-savvy generations, that not everything that looks like an authentic video is in fact a real video may be the first line of defense against the proliferation of deepfakes.
We currently see that mainstream media outlets across the political spectrum are vilified as being producers of fake news. Labels put on deepfake videos warning the video may not be real are unlikely to change the minds of those who want to believe something is, in fact, real. And even for well-intended people, telling the difference between a fake and a real video, or even attributing a real picture to the correct event remains a challenge. There is no ingrained habit of checking sources for online articles, pictures, or videos. Teaching digital awareness may create such a habit over time.
United Nations Secretary-General António Guterres recently expressed his concern about the growing number of protests worldwide. In his speech on 25 October he said “it is clear that there is a growing deficit in trust between people and political establishments, and rising threats to the social contract.” Technologies like deepfakes are increasing the deficit in trust, but they are merely exploiting an already precarious balance. So while the law, Codes of Conduct, digital awareness, and technical means to detect deepfakes are all pieces of the puzzle to help bring back the equilibrium in the marketplace of ideas, let us not let politicians and their parties get off the hook scot-free. They were elected and ultimately hold the key to preserving our liberal democracies. And when they try to limit that democracy, regardless of the tactics or proxies they use, we must stand firm in our rejection of their demagoguery and exploitation of the system if we want to preserve it.
Agnes E. Venema is an a Marie Curie Early Stage Researcher at the Romanian “Mihai Viteazul” National Intelligence Academy and the University of Malta. Her field of research is Intelligence and National Security, and specifically focuses on the intersection between intelligence, technology, and law.
The information and views set out in this article are those of the author and do not necessarily reflect the views of the publisher nor the research funding body.
Chesney, Robert, and Danielle Citron. ‘Deepfakes and the New Disinformation War: The Coming Age of Post-Truth Geopolitics’. Foreign Affairs, February 2019.
Chesney, Robert, and Danielle Keats Citron. ‘Deep Fakes: A Looming Challenge for Privacy, Democracy, and National Security’. SSRN Electronic Journal, 2018. https://doi.org/10.2139/ssrn.3213954.
‘Come Nasce Un Deepfake? Il Dietro Le Quinte Della Rubrica Di Striscia’, 6 March 2020. https://www.striscialanotizia.mediaset.it/news/come-nasce-un-deepfake-il-dietro-le-quinte-della-rubrica-di-striscia_9975.shtml.
‘Deepfake Video of Former Italian PM Matteo Renzi Sparks Debate in Italy’. France 24, 10 August 2019, sec. The Observers. https://observers.france24.com/en/20191008-deepfake-video-former-italian-pm-matteo-renzi-sparks-debate-italy.
‘Deepfakes and Audio-Visual Disinformation’. CDEI Snapshot Series. Centre for Data Ethics and Innovation (CDEI), September 2019.
‘Disinformation and “Fake News”: Final Report’. House of Commons Digital, Culture, Media and Sport Committee, 14 February 2019.
Dwoskin, Elizabeth, Jeanne Whalen, and Regine Cabato. ‘Content Moderators at YouTube, Facebook and Twitter See the Worst of the Web — and Suffer Silently’. Washington Post, 25 July 2019, sec. Technology. https://www.washingtonpost.com/technology/2019/07/25/social-media-companies-are-outsourcing-their-dirty-work-philippines-generation-workers-is-paying-price/.
‘Elections to the European Parliament’. Election Assessment Mission Final Report. Brussels, Belgium: ELECTION-WATCH.EU, 16 September 2019.
‘EU: Russia, China Use “Digital War” to Undermine Democracies’. Associated Press, 30 January 2020. https://apnews.com/ab2e0946183041314846d3419f91de9b.
Hao, Karen. ‘The Biggest Threat of Deepfakes Isn’t the Deepfakes Themselves’, 10 October 2019. https://www.technologyreview.com/s/614526/the-biggest-threat-of-deepfakes-isnt-the-deepfakes-themselves/.
Lee, Alex. ‘How Puny Humans Can Spot Devious Deepfakes’. Wired, 20 October 2019. https://www.wired.co.uk/article/how-to-spot-deepfake-video.
Martens, Bertin, Luis Aguiar, Estrella Gomez-Herrera, and Frank Mueller-Langer. ‘The Digital Transformation of News Media and the Rise of Disinformation and Fake News’. JRC Digital Economy Working Paper 2018-02. JRC Technical Reports. Seville, Spain: Joint Research Centre, European Commission, April 2018. https://ec.europa.eu/jrc/communities/sites/jrccties/files/dewp_201802_digital_transformation_of_news_media_and_the_rise_of_fake_news_final_180418.pdf.
Martins, Felipe. ‘Facebook Lies to Dutch Parliament about Election Manipulation’. Bits Of Freedom, 21 May 2019. https://www.bitsoffreedom.nl/2019/05/21/facebook-lies-to-dutch-parliament-about-election-manipulation/.
Momigliano, Anna. ‘How Russian Propaganda Showed Up in an Italian Murder Trial’. The New York Times, 17 December 2019, Europe edition. https://www.nytimes.com/2019/12/17/world/europe/russia-italy-propaganda.html.
Mounier, Jean-Luc. ‘Inefficace ou mal comprise, la loi contre les “fake news” toujours en question’. France24, 19 June 2019. https://www.france24.com/fr/20190619-france-loi-fake-news-efficacite-promulgation-lrem-macron-fausses-nouvelles-csa?fbclid=IwAR3J1sbmhSDRC3HxOxxeMLZRygaWkzkPo6r9UYUO-F4JXpa_xagu0sljEBw.
‘Putin’s Asymmetric Assault On Democracy In Russia And Europe: Implications For U.S. National Security’. Minority Staff Report, Committee on Foreign Relations, United States Senate. Washington, D.C: U.S. Government Publishing Office, 10 January 2018. http://www.gpoaccess.gov/congress/index.html.
Schipper, Nienke. ‘Nu kan iedereen nepfoto’s en -video’s maken, en dat is best griezelig’. Trouw, 12 September 2019. https://www.trouw.nl/cultuur-media/nu-kan-iedereen-nepfoto-s-en-video-s-maken-en-dat-is-best-griezelig~bf7cf33d/.
Schwirtz, Michael, and Sheera Frenkel. ‘In Ukraine, Russia Tests a New Facebook Tactic in Election Tampering’. New York Times. 29 March 2019. https://www.nytimes.com/2019/03/29/world/europe/ukraine-russia-election-tampering-propaganda.html.
Shao, Chengcheng, Giovanni Luca Ciampaglia, Onur Varol, Kai-Cheng Yang, Alessandro Flammini, and Filippo Menczer. ‘The Spread of Low-Credibility Content by Social Bots’. Nature Communications 9, no. 1 (December 2018): 4787. https://doi.org/10.1038/s41467-018-06930-7.
Tisdall, Simon. ‘Result of Macedonia’s Referendum Is Another Victory for Russia’. The Guardian, 1 October 2018, International edition edition. https://www.theguardian.com/world/2018/oct/01/result-of-macedonia-referendum-is-another-victory-for-russia.
UN / GUTERRES WORLD PROTESTS. New York, NY: UN Audiovisual Library, 2019. https://www.unmultimedia.org/avlibrary/asset/2485/2485406/.
Van de Weghe, Tom, and Judit Verstraete. ‘Facebook gaat deepfakevideo’s verwijderen: “Met die filmpjes kun je politici om het even wat laten zeggen”’. Public broadcaster. VRT NWS. Accessed 28 January 2019. https://www.vrt.be/vrtnws/nl/2020/01/07/deep-fake-facebook/.
Von Der Burchard, Hans. ‘Belgian Socialist Party Circulates “Deep Fake” Donald Trump Video’. Politico Europe. Accessed 28 January 2019. https://www.politico.eu/article/spa-donald-trump-belgium-paris-climate-agreement-belgian-socialist-party-circulates-deep-fake-trump-video/.
 Karen Hao, ‘The Biggest Threat of Deepfakes Isn’t the Deepfakes Themselves’, 10 October 2019, https://www.technologyreview.com/s/614526/the-biggest-threat-of-deepfakes-isnt-the-deepfakes-themselves/.
 ‘EU: Russia, China Use “Digital War” to Undermine Democracies’, Associated Press, 30 January 2020, https://apnews.com/ab2e0946183041314846d3419f91de9b.
 ‘Deepfake Video of Former Italian PM Matteo Renzi Sparks Debate in Italy’, France 24, 10 August 2019, sec. The Observers, https://observers.france24.com/en/20191008-deepfake-video-former-italian-pm-matteo-renzi-sparks-debate-italy.
 ‘Deepfake Video of Former Italian PM Matteo Renzi Sparks Debate in Italy’.
 ‘Come Nasce Un Deepfake? Il Dietro Le Quinte Della Rubrica Di Striscia’, 6 March 2020, https://www.striscialanotizia.mediaset.it/news/come-nasce-un-deepfake-il-dietro-le-quinte-della-rubrica-di-striscia_9975.shtml.
 ‘Deepfake Video of Former Italian PM Matteo Renzi Sparks Debate in Italy’.
 Robert Chesney and Danielle Keats Citron, ‘Deep Fakes: A Looming Challenge for Privacy, Democracy, and National Security’, SSRN Electronic Journal, 2018, 28, https://doi.org/10.2139/ssrn.3213954.
 Nienke Schipper, ‘Nu kan iedereen nepfoto’s en -video’s maken, en dat is best griezelig’, Trouw, 12 September 2019, https://www.trouw.nl/cultuur-media/nu-kan-iedereen-nepfoto-s-en-video-s-maken-en-dat-is-best-griezelig~bf7cf33d/.
 Anna Momigliano, ‘How Russian Propaganda Showed Up in an Italian Murder Trial’, The New York Times, 17 December 2019, Europe edition, https://www.nytimes.com/2019/12/17/world/europe/russia-italy-propaganda.html.
 Simon Tisdall, ‘Result of Macedonia’s Referendum Is Another Victory for Russia’, The Guardian, 1 October 2018, International edition edition, https://www.theguardian.com/world/2018/oct/01/result-of-macedonia-referendum-is-another-victory-for-russia.
 Michael Schwirtz and Sheera Frenkel, ‘In Ukraine, Russia Tests a New Facebook Tactic in Election Tampering’, New York Times, 29 March 2019, https://www.nytimes.com/2019/03/29/world/europe/ukraine-russia-election-tampering-propaganda.html.
 Schwirtz and Frenkel.
 Lee, ‘How Puny Humans Can Spot Devious Deepfakes’.
 Troll Factories agglomerations of fake online profiles pushing a certain message out through various social media networks, often employing circular ‘reporting’ tactics. They are known to share photo shopped images and doctored videos.
 Chesney and Citron, ‘Deepfakes and the New Disinformation War: The Coming Age of Post-Truth Geopolitics’, 148.
 ‘Deepfakes and Audio-Visual Disinformation’, CDEI Snapshot Series (Centre for Data Ethics and Innovation (CDEI), September 2019), 13.
 ‘Deepfakes and Audio-Visual Disinformation’, 13.
 ‘Disinformation and “Fake News”: Final Report’ (House of Commons Digital, Culture, Media and Sport Committee, 14 February 2019), 13.
 Jean-Luc Mounier, ‘Inefficace ou mal comprise, la loi contre les “fake news” toujours en question’, France24, 19 June 2019, https://www.france24.com/fr/20190619-france-loi-fake-news-efficacite-promulgation-lrem-macron-fausses-nouvelles-csa?fbclid=IwAR3J1sbmhSDRC3HxOxxeMLZRygaWkzkPo6r9UYUO-F4JXpa_xagu0sljEBw.
 Chesney and Citron, ‘Deep Fakes’, 34–36.
 Chengcheng Shao et al., ‘The Spread of Low-Credibility Content by Social Bots’, Nature Communications 9, no. 1 (December 2018): 4787, https://doi.org/10.1038/s41467-018-06930-7.
 See for example: Bertin Martens et al., ‘The Digital Transformation of News Media and the Rise of Disinformation and Fake News’, JRC Digital Economy Working Paper 2018-02, JRC Technical Reports (Seville, Spain: Joint Research Centre, European Commission, April 2018), https://ec.europa.eu/jrc/communities/sites/jrccties/files/dewp_201802_digital_transformation_of_news_media_and_the_rise_of_fake_news_final_180418.pdf.
 Elizabeth Dwoskin, Jeanne Whalen, and Regine Cabato, ‘Content Moderators at YouTube, Facebook and Twitter See the Worst of the Web — and Suffer Silently’, Washington Post, 25 July 2019, sec. Technology, https://www.washingtonpost.com/technology/2019/07/25/social-media-companies-are-outsourcing-their-dirty-work-philippines-generation-workers-is-paying-price/.
 See for example, ‘Disinformation and “Fake News”: Final Report’.
 ‘Putin’s Asymmetric Assault On Democracy In Russia And Europe: Implications For U.S. National Security’, Minority Staff Report, Committee on Foreign Relations, United States Senate (Washington, D.C: U.S. Government Publishing Office, 10 January 2018), 116, http://www.gpoaccess.gov/congress/index.html.
 ‘Elections to the European Parliament’, Election Assessment Mission Final Report (Brussels, Belgium: ELECTION-WATCH.EU, 16 September 2019), 22.
 ‘Elections to the European Parliament’, 23.
 ‘Elections to the European Parliament’, 23.
 Felipe Martins, ‘Facebook Lies to Dutch Parliament about Election Manipulation’, Bits Of Freedom, 21 May 2019, https://www.bitsoffreedom.nl/2019/05/21/facebook-lies-to-dutch-parliament-about-election-manipulation/.
 ‘Elections to the European Parliament’, 24–25.
 Schipper, ‘Nu kan iedereen nepfoto’s en -video’s maken, en dat is best griezelig’.
 Hans Von Der Burchard, ‘Belgian Socialist Party Circulates “Deep Fake” Donald Trump Video’, Politico Europe, accessed 28 January 2019, https://www.politico.eu/article/spa-donald-trump-belgium-paris-climate-agreement-belgian-socialist-party-circulates-deep-fake-trump-video/.
 Tom Van de Weghe and Judit Verstraete, ‘Facebook gaat deepfakevideo’s verwijderen: “Met die filmpjes kun je politici om het even wat laten zeggen”’, public broadcaster, VRT NWS, accessed 28 January 2019, https://www.vrt.be/vrtnws/nl/2020/01/07/deep-fake-facebook/.
 Van de Weghe and Verstraete.
 UN / GUTERRES WORLD PROTESTS (New York, NY: UN Audiovisual Library, 2019), https://www.unmultimedia.org/avlibrary/asset/2485/2485406/.
 ‘Disinformation and “Fake News”: Final Report’, 74.
Photo: Government of Gabon
Published on August 4, 2020.