The scary world of Deepfakes and how to spot one

Deepfakes are machine learning-enabled software programs that add or replace a person’s face on a video of another person, in real-time. Deepfake refers to the use of Generative Adversarial Networks (GANs) with the intent to create videos that can be used for misleading or deceiving purposes.

Generally, deepfake users will take footage of people from footage that is available on public platforms like YouTube, or movies and then synthesize the individual’s voice using speech synthesis technology. 

The video will then be posted on social media or on porn sites, in order to deceive others into thinking it is authentic. The majority of the victims are female celebrities that have their Blonde Pussy exposed online in fake porn videos.

The deepfakes phenomenon was largely brought into the spotlight when celebrities including Emma Watson and Taylor Swift began having deepfake images of themselves spread around the internet (mostly on porn sites). 

How to identify a deepfake

Blinking: people usually blink once every 2-8 seconds and each blink lasts between 1 and 4 tenths of a second. In this type of fake video, the person usually blinks much less than a real person. 

Face, neck, and body: most deepfakes only change the person’s face, so changes to the body can only be changed with great effort. Therefore, the body of the person whose face is replaced does not usually match the facial expression in posture. Similarly, tattoos, muscles, or different skin colors may indicate that it is a fake video

Short clips: Most fake videos are usually very short in content, only a few seconds. There is no reason for the content of the recording to be so short, something that also indicates that it is a ‘deepfake’.

Source of the recording: it is important to find the person who first shared the fake video. The context is important to know if it is a real video or not. 

Sound recording: if the sound is not present or does not match the image, it is a fake.

Details: it is useful to let it play at half speed, as it is easier to see details or sudden changes in the image.

Inside of the mouth: this detail also indicates the falsification of the video. For example, some blurring of the inside of the mouth is another sign that it could be a fake image.