Given the current unprecedented level of information warfare waged by the Collective West against those countries that do not agree to live under the outdated rules of the current global world order, we are seeing an increasing focus on attempts to expand the range of tools available by using new technologies based on artificial intelligence.
One of the notable phenomena of recent years in this area has been a boom in the creation of deepfakes, which are defined as “realistic substitution of photo, audio and video materials with the help of artificial intelligence.” The term itself is a combination of “deep” learning (machine learning by neural networks) and “fake.” At the heart of deepfake technology, simply put, is the collection and analysis of a large number of photographs and video clips of a particular person, with which the neural network builds an algorithm of how the person might look and move. Among the main types of deepfake are face substitution, in which the face of one person is superimposed on the face of another, speech synthesis based on the voice of a real person, and the creation of entirely computer generated videos with substituted faces and synthesized speech. Then a new video is created in which a character with the face of a real person speaks and acts in the way that the creator of the video wants.
Today, new capabilities are emerging for image generation, voice processing, and combining computer graphics with real video. The areas in which this new type of content is applied are rapidly expanding- it is now being used in advertising, marketing, entertainment videos, computer games, science, education, journalism, painting, cinema, the fashion industry, design and architecture.
Deepfakes are being created for entertainment purposes, including to recreate the images of deceased celebrities, “rejuvenate” current actors and play a wide variety of pranks and jokes, etc., and these are spreading rapidly via social networks. For example, in a documentary film created using deepfake technology, the great surrealist painter Salvador Dali has been “resurrected” and now “communicates” with visitors to the museum named after him in Saint Petersburg (USA). And at the end of 2019, a video of a deepfake “alternative Christmas speech by Queen Elizabeth II” caused widespread outrage. The picture at the head of this article shows a frame from a video of what appears to be the well-known movie star Tom Cruise.
The fact that an increasing proportion of materials is now being generated in this way is an important indicator of how this technology has grown. It has been predicted that in the near future artificial intelligence will be used to create a significant portion of all digital content on the Internet.
Given this background, the proliferation of deepfake materials in the turbulent sphere of modern global politics no longer seems harmless or amusing. After all, the fake politicians or celebrities appearing in materials created with this technology say things the real people never said, and do things they never did. As a result there is a real danger of social networks being used as a channel for spreading fake news, slandering opponents, manipulating public opinion and committing other illegal acts that may have far-reaching political consequences.
Russian lawmakers are justifiably concerned about the dangerous consequences of the uncontrolled dissemination of deepfake materials. The State Duma of the Russian Federation is currently considering a bill designed to create a strict legal framework to regulate this type of activity in Russia.
Western politicians and political operatives appear to be taking a diametrically opposite view of the deepfake issue – namely that a skillfully used “fake” can have successful results. For example, in unfolding US presidential election campaign, which promises to be one of the dirtiest in the country’s history, there have already been many examples of the creation and dissemination of openly provocative deepfakes. For example, an audio fake has been discovered in which the voice of Joe Biden “strongly encourages” voters to abstain from voting in the Democratic Party primaries in New Hampshire.
A report by the Washington-based Center for Strategic and International Studies, one of the country’s most respected think tanks, analyzes the role played by deepfakes in US political life and its potential for future development.
Clearly aware of the fact that the production and distribution of deepfakes amounts to the use of “technologies calculated to mislead,” the authors of the study find it necessary to adduce arguments to justify the West’s use of such practices, even if these are clearly equally misleading. As one might expect, the American experts see a “Russian connection” as one of the reasons for the information warfare using deepfakes that is unfolding in global media. However, their “arguments” appear, to say the least, rather unconvincing. The allegedly fake materials cited in the report include videos purporting to show Ukrainian President Zelensky against a background of Nazi swastikas and greeting participants in a gay parade in Kiev. Referring to these materials, the authors simply state that these materials are “widely believed to have been created and disseminated by Russia.”
As a “countermeasure,” the document justifies the use by “democratic countries of the Collective West” of offensive deepfake tactics.
Without waiting for official approval of their initiatives, the center’s staff, with the involvement of “external experts,” are already working out hypothetical scenarios for possible operations based on the use of deepfakes. Here is a sample of one of them. “Country X … has a government that is not friendly to Western countries… In the aftermath of recent elections, Country X is experiencing political instability, with many protests and moderate and pro-Western leaders credibly contending the election was stolen from them. In the midst of ongoing instability, intelligence agencies propose to use deepfakes showing the leader of Country X thanking a Western diplomat for sending him money, receiving a payoff from a wealthy national of Country X, laughing at the deaths of citizens of Country X in protests, and ordering ballot boxes to be stuffed with fraudulent ballots. The intelligence agencies hope this will discredit him and contribute to his downfall.” According to their developers, the implementation of the proposed options could “discredit him [the leader of Country X] and contribute to his downfall.”
On the practical level, the White House is advised to elevate the use of deepfakes to the level of “a specific field of national security operations.” The authors propose that an interagency working group composed of representatives from the US Department of Defense, Department of State, Department of Justice, Department of the Treasury, and the CIA be formed to coordinate these activities within the administration. The report stresses the need to carefully consider the double-edged nature of the use of deepfakes, which consists of both great opportunities if successful and serious consequences if exposed. Therefore, in making decisions on operations in this area, the authors of the report propose that the US president should have the last word.
In view of the novelty and sensitivity of the problem of the use of deepfakes in politics, representatives of the US expert community are recommended to start analyzing the various components of this issue. As priority issues for discussion, the authors propose that assessments of the effectiveness of such actions be developed, the target audience be identified, a study of the legal implications of the use of deepfakes be performed etc.
The authors of the document assume that the development of AI technologies provides “an unparalleled opportunity for state actors to use deepfakes for national security purposes.” They believe that it will not be long before the US and other “major democracies” move to widespread use of such designs “to mislead for various legitimate purposes in the highest interests,” with a small caveat “if they have not already done so.”
Leonid Gladchenko, expert political specialist and member of the Analytica Association, exclusively for the Internet magazine “New Eastern Outlook”
