Deepfakes: Historical Context, Advantages, and Challenges

Deepfakes
Abstract
Deepfake technology is a new trend that started in 2017. It entails using artificial intelligence (AI) to create or alter videos, audios and images. It is a new trend that social media strategists can use to advertise various products. Among the benefits of deepfake technology is that it is cheap because it if free through apps like Zao, It is also reliable and maintain the message’s authenticity. Its challenges are that people may use the app maliciously to revenge people and damage their images. An individual or organization needs to get the best AI algorithm to create deepfakes. When implementing this technology, the app or algorithm used should be licensed to avoid copyright or patent breach. A company needs someone who understands how to make deepfakes and since it is not a complicated process, the present information technology specialists can be trained to do it perfectly.
Deepfakes
Introduction
Deepfake technology is a new trend that started in 2017. Indeed, it involves using artificial intelligence (AI) to make and alter videos, audio, or images. In other words, deepfake technology can be used to audio, images, or videos digitally to spread information. Social media strategists or managers can benefit from this new trend by helping organizations and people to present themselves on social media appropriately. These communication practitioners analyze how individuals communicate on social networking platforms, and they advise them accordingly. However, they should understand how to overcome the challenges posed by deepfake technology. The paper focuses on deepfake technology and how social media strategists can use it to promote individual or organization brand name and reputation creation and maintenance.
The primary readers of this paper are social media strategists who want to understand how to use social media platforms, such as TikTok, to promote new product launches. These target audiences will benefit the most by knowing how they can use the deepfake technology. Social media strategies work in various communication fields, including political, public relations, corporate branding, activism, healthcare, education, transport, fashion, and personal branding. The use of deepfake video technology is rapidly increasing. For instance, there exists a face-swapping app known as Zao, which allows users to swap faces with celebrities to create a deepfake video within a few seconds (Rees, 2019). As such, the deep generative modeling enables users to generate duplicate actual faces to create impressively true-to-life images, audios, and videos.
Historical Context
Deepfake technology is a relatively new as it begun in 2017. Deepfakes emanate from using AI to merge, superimpose, combine, and replace videos and images to come up with fake videos that look authentic. A deepfake technology can be used to generate a political, humorous, or pornographic video of an individual saying something without the consent of the one whose voice and the image are involved (Westerlund, 2019). For instance, in a 2019 CNN Business article "When Seeing is no Longer Believing," there is a deepfake video of the former president of the United States of America (USA) saying that "We are entering an era which our enemies can make it look like anyone is saying anything at any point in time" (CNN Business, 2019). In other words, with deepfake technology, it is possible to take the face and voice of someone to create a fake video where the person seen talking was never involved.
As mentioned in the beginning, the first deepfake video was uploaded in 2017 by a Reddit user. The creator posted pornographic video clips that swapped the faces of different celebrities, including Scarlett Johansson, Gal Gadot, and Taylor Swift (Sample, 2020). In reality, it takes a few seconds to swap faces. Notably, a person runs thousands of faces of two individuals in an AI algorithm known as the encoder. The encoder learns and finds the similarities between the two faces and compresses the images showing the commonly shared features (Sample, 2020). A decoder is then used to recover the faces based on the one that the creator wants to use in a deepfake. Based on Deeptrace, in September 2019, there were around 15,000 deepfake videos on the Internet, and about 96% were pornographic videos, where 99% of them swapped the faces of female actors for those of celebrities (Sample, 2020). As a result, deepfake technology is being increasingly used to achieve the right and wrong outcomes.
Benefits to social media strategists
Social media strategists can use deepfake technologies in various ways to promote new products on social networking platforms. First, advertisers can use deepfakes to create a multilingual advert from a single influencer. Here, the influencer will appear to be advertising in many languages. For example, in 2019, an international malaria public awareness campaign video was created featuring David Beckham. The video overcame language barriers by using voice and visual-altering technology or deepfake, making Beckham appear multilingual (Westerlund, 2019).
Secondly, the deepfake technology can be used to translate a speech and alter the mouth and facial movements simultaneously to facilitate eye contact and make the person speaking use different languages. This is a great feature for social media strategists because they can use it to reach people who speak different languages, appealing both to audio and visual aspects. In that light, social media strategists can use this new technology to reach the target population by creating ads that speak a similar language to those who are targeted (Kerner & Risse, 2020).
Thirdly, deepfake technology retains the message's originality or authenticity. When social media strategists create an advert using deepfake technology in numerous languages, the original message is always retained throughout the various adverts. This is because the technology translates the orig...