Deepfake materials are becoming more and more widespread and are steadily gaining ground in media and in everyday discourse. Many videos published in the recent years shocked or misled the public and unveiled the potentials of the technology and its effects on democratic societies.
Deepfake is considered a synthetic media that replaces another person’s face or voice with someone else (https://dictionary.cambridge.org/dictionary/english/deepfake, 2022.06.26.). Naturally, it does not necessarily mean replacing a real person with another real actor, and the deepfakes can also animate complex scenes with realistic acts or remarks made by the given person, who in real life has never been associated with such acts or remarks in the first place. Deepfakes first made their appearance in late 2017 and were first used to create the fake impression of celebrities appearing in pornographic videos. The technology, however, quickly found its way into many other domains from politics to media and entertainment and a wide number of other services.
Examples from the recent past includes a deepfake video created by artists Bill Poster and Daniel Howe with the help of an advertising company, featuring Facebook founder Marc Zuckerberg making a threatening announcement about the amount of data they control (https://www.youtube.com/watch?v=Ox6L47Da0RY, 2022.06.26.) or an extremely realistic video about Tom Cruise created by Belgian graphic artist, Chris Ume (https://www.youtube.com/watch?v=wq-kmFCrF5Q, 2022.06.26.).
The common trait of these videos is that they are both extremely realistic and make a lot of people believe that they are genuine, which could lead to a declining trust in media and to the mass spread of fake information aimed at discrediting public figures or influencing economic trends or social movements. Examples from politics include a deepfake video of Ukraine’s president telling the people to lay down arms or the Russian president declaring peace (Jane Wakefield, Deepfake presidents used in Russia-Ukraine war, BBC Technology, 2022.03.18, https://www.bbc.com/news/technology-60780142, 2022.06.26.).
It is highlighted with regard to the above that deepfake detection tools have already been deployed by major social media and technology companies and experts also continue to fight against the spread malicious deepfakes. This race, however, seems never ending, since misleading materials continue to be more and more common and convincing at the same time. At one point we can possibly arrive to a future, where synthetic media would appear to be indistinguishable from real materials, and this future is not necessarily far away.
With regard to the above, media literacy and analyzing news with a critical viewpoint are becoming increasingly important. Education programs in the upcoming years should introduce courses or materials about critical reading and techniques to recognize potential deepfakes or other harmful contents into their curricula, and journalists have to act with more caution as well. It is beyond doubt that detection tools and experts will continue to have a key role, however, they cannot be the sole answer for the threat of hyper-realistic, malicious deepfakes. It is the careful reader, who can be the perfect balance and help democracies fight against disinformation.
In addition to the above, it also needs to be mentioned that deepfakes cannot always be regarded as threats or harmful materials. They also have the potential to revolutionize art and entertainment and to serve as a way of expression and communication. One day, we can be part of our favorite movie, for example, or use the technology to relive joyful moments with our family and friends.