Have you ever imagined seeing your face in a totally random video? ¿Or hear your own voice saying things you would never say? Well this is possible thanks to a technology called deepfake, which allows this manipulation, especially in videos, quite convincingly. Here are more details on what a deepfake is and why it is important to be aware of this technique.
You Can Join Our Social Platform S32 Social Only For Hackies And Techies And Share Your Knowledge And Make Community More Bigger, Thank You.
WHAT IS DEEPFAKE
Deepfake uses Artificial Intelligence to change people’s faces on videos, synchronize lip movements, expressions, and other details, with sometimes impressive and very compelling results.
This tendency to change faces in videos is not new. However, in December 2017, a Reddit user named deepfakes went one step further. Using Open Source Artificial Intelligence and Automatic Learning tools such as Keras and TensorFlow (the latter from Google), created an algorithm to train a Neural Network to map the face of one person in the body of another, frame by frame.
Instead of relying on manual editing as before, the user, through the tool (called Deepfake), only needs one source to recognize the facial model of the “ victim ”, map the structure of the target head and overlay. The software can adjust the mobility of the original video to the new face, including facial expressions and lip movements.
Deepfake initially required advanced user knowledge. However, when others created applications and systems capable of automating the entire process, the tool began to be used massively for malicious purposes: Numerous adult videos edited with the software soon began to appear, using the faces of actresses and artists.
Deepfake’s “ fun ” was not limited to adult videos; There are a number of creations that put celebrity faces in movies where they were not present. However, the possibilities of using the software are numerous and, it should be noted, harmful. Even Mark Zuckerberg was the subject of a fake video.
Some examples of easily accessible deepfake tools are: the application Reface, the sites DeepSwapand DeepFakes Web, and the FaceApp and DeepFaceLab programs.
Importantly, deepfake is different from shallowfake. The premise is similar, but the difference lies in the quality. In shallowfake, the quality of handling is inferior, making it easier to identify.
DEEPFAKE TYPES
Although deepfake’s main focus is exchange of faces in videos, It would be a mistake to think that practice is limited to that. The technique is also used for manipulation of audios, creating recordings that simulate the voice of a specific person, a type of deepfake easily shareable in messaging applications such as WhatsApp.
Also, we are already meeting textual deepfakes, with machines that generate writings using artificial intelligence; deepfakes on social networks, for creating false profiles on the internet; and deepfakes in real time, where it is possible to change the face in live broadcasts, a technique allowed by software such as DeepFaceLive, for example.
DEEPFAKES POTENTIAL RISKS
Although the main objective of the community around this tool is public people, nothing prevents someone with bad intentions from taking a video of an ordinary person and placing him in an embarrassing situation.
It’s not just about edited revenge videos: it’s possible, for example, to create a false moment among friends to generate an alibi, testimony to persuade others that you support a certain point of view or even a crime situation to incriminate you. It all depends on the creativity of the editor and how many of your videos are readily available on the internet.
Since the average video quality is not very high due to hardware demands (at least one GeForce GTX 1050 is required for acceptable results), These creations can be convincing for someone unsuspecting, who might think that it is a real video, even if it has some flaw.
HOW TO IDENTIFY A DEEPFAKE
When receiving a dubious video, stop and watch it several times, at all speeds, to make it easier to identify faults.
It is also important to pay attention to the person’s face in the video. Observe skin, eyes, eyebrows, and mouth, for example. Through these observations, you will be able to identify some anomaly, such as slower blinking, very smooth or wrinkled skin, among other details.
Finally, investigate information about the content received (be it video, audio or text). You may find related news on the internet. In addition, try to find another version of the publication in order to compare statements and gestures.
In short, pay maximum attention to all the details.
STEPS TO TAKE
To prevent your content from being used in a deepfake, The most sensible thing is to avoid sharing personal videos with unknown people or not publishing them on social networks publicly to hinder the editor’s work.
In general, society is still adapting to deepfake because it is a new technology and, if used incorrectly, it can be extremely damaging. New legal mechanisms are expected to be approved to combat this practice and protect people. Therefore, if you have been a victim, collect all the necessary information and search for a lawyer or the Public Defender as soon as possible.
In the case of videos and audios received, avoid sharing them without following the instructions to identify a deepfake. This attitude is extremely important to prevent the spread of false information and inappropriate content.
¿IS IT ILLEGAL TO CREATE DEEPFAKE?
Actually, artificial intelligence and deep learning technologies are not a problem, as they can be used to create harmless videos. However, the word deepfake is generally associated with negative situations that pose risks to individuals, social groups, and even democracy.
About System32 Social
You Can Join Our Social Platform S32 Social Only For Hackies And Techies And Share Your Knowledge And Make Community More Bigger
- Posting System
- MarketPlace For Selling Buying Tools
- Forum (0day Exploits, Pentesting Tools, etc.)
- Messaging System
Visit Now: https://system32.in
