As technology progresses in the twenty-first century, numerous technologies have been developed to make people’s lives easier. But, as with any advancement, not all of it was intended to be beneficial for humanity. Deep fake videos, which use artificial intelligence (AI) called deep learning to generate photos of fake events, are one of the things to watch out for in today’s society.
As technology progresses in the twenty-first century, numerous technologies have been developed to make people’s lives easier. But, as with any advancement, not all of it was intended to be beneficial. Deep fake videos, which use artificial intelligence (AI) called deep learning to generate photos of fake events, are one of the things to watch out for in today’s society.
Making of deepfake videos
The most popular form of deepfake media that’s circulating the internet is videos. It can push one person’s head into another person’s body or make them say something else that they won’t know themselves. Deepfakes are simple to make and take just a few steps, so anyone with the right resources can do it. Thousands of photos of the two people are fed into an encoder, which is an AI algorithm. The encoder then searches for resemblance points between the two faces, reducing them to their shared standard features, compressing the two images while generating a new one.
The faces are then retrieved from the compressed images using a second algorithm called a decoder. When two decoders are used, the result shows the first person’s face and the gestures and orientation of the other. The process must be replicated on each frame to produce a compelling video. This type of technology or knowledge isn’t exclusive to a single company. Several apps have been created to perform such procedures, allowing almost everyone to make deepfakes according to their preferences.
Threats associated with deepfakes and how to avoid them
Deepfakes can appear to be harmless, but they can be harmful to the untrained eye. It has the potential to spread malicious lies, lower stock prices, and trigger fear by undermining public confidence.
As technology advances, detecting deepfake videos becomes more difficult. There are, however, several low-quality deepfakes that are easier to discern. Lip-synching off the mark, patchy skin tone, and flickering around the edges of transposed faces are all telltale signs.
You can also find some clues in the finer details, such as the hair, where poor deepfakes leave strands visible on the fringe. Poorly rendered jewelry and teeth, as well as odd lighting effects that display inconsistent illuminations and reflections on the iris, maybe a dead giveaway of deepfakes.
Governments, universities, and software companies are currently funding research into how to detect deep fakes. However, for the time being, the public is urged to remain cautious and do their investigation to ascertain the facts behind widely circulated media.