Deepfake technology creates fake videos that appear to be real from artificial intelligence (AI). In general, these contents show celebrities and politicians doing or saying something that did not happen. The term came about when a Reddit user with that name posted pornographic videos featuring the faces of celebrities. Since then, the montage of videos has already generated satirical content and even criminals, when they involve false identity and information that is not true. Public figures such as Barack Obama, former president of the United States, Mark Zuckerberg, one of the founders of Facebook and Donald Trump, current US president, were targets of deepfake. In addition, actress Gal Gadot and singer Taylor Swift had their faces associated with pornographic videos produced using this technology.
READ: Eight shameful celebrities on the internet in 2019; Look
To produce these videos, software uses algorithms with the original face and voice of the person who appears in the final product. Therefore, it is easier to produce this content with images of public figures. However, the exposure of ordinary people to social media also makes it easy for anyone to be a victim of deepfake technology. Next, dnetc selected five impacts – positive or negative – that deepfake can cause in the coming years.
Gal Gadot is among the artists who were victims of deepfake Photo: Pond5
Want to buy cell phones, TV and other discounted products? Meet Compare dnetc
Like fake news, deepfake technology can convince a crowd that the content is displayed on real video. Although there are clear differences between a real video and one made from this feature (such as skin that resembles wax, the lack of synchrony between audio and mouth movement and the rare times that the character blinks), some people believe in veracity of the content.
So far, AI does not produce 3D content and stretches the character's face to disguise the movements. Because of this, the human eye can perceive that the ends of the face are quite marked. Thinking about combating false information, fact checking should verify, in the future, the veracity of these videos.
In fact, the use of this technology to shape public opinion is already expected in the 2020 elections in the United States. In order to prevent these contents from spreading, the US Congress started to demand that the videos be properly classified as deepfake, with a watermark that identifies them.
Video with Barack Obama speech looks real, but words are from director Jordan Peele Photo: Reproduction / BuzzFeed
2. More accessible technology
Although some software covers to remove the watermark from tampered videos, free programs are already used to produce deepfakes. However, a good video card is still needed for the content to be processed more quickly, which is not accessible to anyone. With the updates of artificial intelligence in the coming years, the use of this software will be easier and more accessible, which will enable anyone to produce this content.
According to a survey by Deep Trace, there are more than 14,000 deepfakes with pornographic content on the web. In 96% of cases, the face of women used in products. In this way, these fake videos can be used to promote revenge pornography and the lynching of these women.
Currently, two techniques are used to produce the videos. The first uses real pornography images and only changes the face of actresses to that of a known person. The second, known as Deepnude, uses the ordinary photo of a person, removes his clothes and creates the pornographic image.
4. Commitment to facial recognition technology
Recreating anyone's face from deepfake technology is not too complicated. From this, facial recognition used to unlock devices and enter applications is also compromised. The more realistic the displayed image, the better the devices process recognition and unlock computers and cell phones. Thus, companies must invest in technologies that block these images.
5. Treatment of depression
Although deepfake's algorithms are widely used to promote crimes on the Internet, the technology can also be used to treat diseases such as depression. With the use of someone else's image, the patient who is not comfortable exposing his problem can report it to a psychologist or psychiatrist without having to identify himself. In addition, depression can also be recognized by AI from the person's tone of voice.
YouTube does not upload the videos: why? See tips in the dnetc forum
How to record Android phone screen on video
