As long as it’s on the air, the internet encyclopedia Wikipedia, which will never end up being updated by online users, shows the “deepfake and synthetic media effect” among 7 mega Trends. Wikipedia, which defines the Mega trend as trends with global impact, emphasizes that some of them are global threats. One of the mega-trends considered a global threat is undoubtedly “deepfake and synthetic media”.
7 mega trend’s close relationship with global threats
The mega trends that Wikipedia ranks are determined by sources from 2017-2020, which also make up deepfake’s history. The mega-trends that humanity is most concerned with reveal the global whole that actually defines the cross-section of time in which we live. They all interact with each other, each giving way to the other. One comes across as the cause or consequence of the other. “Microbiomes and synthetic biology”, for example, from today’s global threats to “climate change and the depletion of Natural Resources”, is also on Wikipedia’s mega-trend list. Can we think of these independently of the mega-trends of “technological progress” and “rapid global urbanization”? Isn’t our global nightmare, the Covid-19 pandemic, already a result of this process?
Of the viruses that can infect the cells of all kinds of creatures, do you know which ones are considered useful? Those such as “latent viruses” that have the ability to remain dormant within the cell. Because they can increase immunity to certain bacteria, such as the “yersinia pestis bacterium”, which has led to deadly plague pandemics throughout history. Deepfake may also become more useful than it is today when it begins to strengthen our immunity to global threats, such as disinformation, by reining in malicious use. And to control it, we need more time. However, like synthetic biology, synthetic media has turned into an uncontrolled frenzy. Deepfake is also constantly mutating in the conjuncture shaped by these 7 mega Trends. Like a menacing virus whose unknowns cannot be dealt with, it continues to spread rapidly.
Lord of the Deepfakes
“The Lord of the Rings” is the immortal work of John Ronald Reuel Tolkie, a German-born English writer, poet, philologist and professor at the University of Oxford, specializing in Anglo-Saxon language and literature. The author’s famous trilogy of fantasy novels, the first of which was published in 1954. The trilogy of the same name as the novel moved to the white screen almost half a century later, directed by Peter Jakson. “Brotherhood of the ring” was the first film of this trilogy to be released in 2001. This first film in the trilogy was a fictional medieval novel. The film was about the struggle to capture unique power to destroy a talismanic ring that would lead the world into bloody wars. The characters, adapted from the novel, overcame the dangers with the unity of power symbolized as the “Brotherhood of the ring”, and led the audience on a mystical adventure.
The “Deepfake Brotherhood” of deepfake developers, whose numbers are growing rapidly, is also giving impetus to the deepfake adventure of the online world, in fact with a similar interaction. But this time we are faced with a global solidarity that aims not to destroy the talisman strong enough to determine the fate of the world, but to share and reproduce it. Although it may not seem organized, the “Deepfake Brotherhood” refers to a joint attempt to bring fake synthetic media to a hyper-realistic quality that can be considered the “master of the deepfakes.” He reveals a joint, anonymous struggle. Fantastic stories, just like Jules Verne’s “20 Thousand Leagues under the Seas”, convey predictions of human intelligence that can be considered a little mystical. They touch our imagination, independent of time and space, while informing us of the future. Like the “Brotherhood of the Rings,” which symbolizes medieval togetherness, the “deepfake Brotherhood” of the digital age draws its power from joint effort.
Deepfake Brotherhood of Generation Y and Z,
When you investigate, you come across countless true stories. Triggering Generation Y and Z, in particular, for deepfake, leads to an avalanche effect. A deepfake humor sharing on social media or a hands-on training video posted by a deepfake volunteer is enough to get many enthusiasts excited.
Grace Windheim, a recent graduate of the university, for example, says that deepfake has entered the field of viral attraction when it was never in her mind. Windheim, who works as a content editor at Kapwing, which offers a video editing software package to its users, not only develops herself in a short time, taking advantage of the resources she has accessed through internet searches on deepfake on which she had no prior knowledge. She also transmits what she has learned and practiced to others through a training video that she has posted on YouTube. Windheim’s deepfake training video on YouTube has been viewed more than 360 thousand times in a month from August 4.
James Vincent, a young technology reporter who has written AI articles for important international publications such as The Independent, The Financial Times and the BBC, also shared his first deepfake experience on 9 September at theverge.com in his post. In his article titled “I learned how to do lip syncing and you can too in just a few hours” Vincent shows special Lip synchronization algorithm wav2lip that is produced by an international team of universities in India and the UK. Vincent has been sharing his first examples of deepfake on Twitter, which he has produced by overcoming his hangouts with the help of a YouTube training video.
As Deepfake turns into “child’s play,” the need for safety increased
Digital synthetic media, produced by machine learning algorithms over artificial neural networks, is about to achieve a production quality that is becoming increasingly difficult to distinguish from the reality with the “share-develop” model. In addition, open source software, educational videos that encourage and facilitate the use of this software, freelance deepfake manufacturers and technology companies that develop deepfake technology and market deepfake products that they produce have enabled all types of access.
Deepfake technologies are becoming so quickly comparable to reality that it attracts the online world like a magnet with its irresistible charm. Even those who come together, influenced by this area of attraction, are creating online deepfake communities. Easy access to Deepfake models and unlimited sharing also turn malicious usage scenarios that will turn into cyber attacks into a major threat. Deepfake makes the need for detection tools acute.
Deepfake guides for deepfake communities
Generation X remembers well. In the early years of the internet and Google has not yet become our online guide, websites that list the most clicked sites were very popular. These sites, which listed the websites that received the most hits in general or categorically with rankings such as the Top 100, Top 500, also increased access to the sites they listed. The content created under the “Deepfake community” image is actually not much different from them. Hosting the emergence of deepfake Reddit.com, also embraces the creation of deepfake libraries again.
Media synthesis, which has been operating on Reddit.com since the beginning of 2018, stands out as one of the most active deepfake community channels with 19 thousand members. In its definition of community, it emphasizes the more recreational use of synthetic media. This platform covers image synthesis, sound synthesis, text Synthesis, style transfer, speech synthesis and more. Media synthesis’s “Resource Library for synthetic media “consists of four parts. In the “image synthesis-processing-development” section, there are interesting mobile and desktop applications such as Impressions, Artbreeder,9GAN, Deepart.io, GANbreeder, Sketch RNN, GauGAN Demo, including summary information and access links. Under the heading “text synthesis”, a total of 5 programs developed for natural language processing, language modeling and text synthesis, such as InferKit, Grover, Inspirbot. In the section “sound synthesis and Music Production”, 6 different sources for sound are transferred, such as Jukebox, AIVA, Lyrebird and MuseNet, which allow you to create or manipulate digital interface (MIDI) files and waveforms of musical instruments. Under the heading “classics”, 7 different options appear, such as WolframTones, Natural Reader and Donjon, along with traditional acquaintances such as Photoshop.
SFW deepfakes community on the other hand, which was founded in the same year and even the same month on Reddit.com as Media Synthesis, has a much different mission. On the platform, where all types of software sharing, including open source ones, and all types of advertising or spam are prohibited, only “business-safe” deepfake shares can be made, as the name suggests. The Platform brings deepfake developers together, especially in the common denominator of avoiding porn deepfake examples.
Deepfake volcano expanding rapidly…
Large-scale volcanoes, described as super-volcanoes, have been known to have the potential for incredible danger, even if they remain silent for thousands of years, leading to the destruction of a continent or climate change on earth when they erupt. Deepfake technology is also expanding and bubbling online like a volcano ready to erupt. Although it can be seen spraying lava from its peak dish, the hot lava has not yet started to flow down the slopes and scorch. But it is certain that the growth in communities through anonymous development-sharing solidarity has led to the rapid expansion of the deepfake volcano. It’s not known when it explodes, but it doesn’t look too far away…