How To Spot Deepfake Videos
AI and machine learning have been some of the most talked-about technologies in the last decade. However, these technologies are also being misused to create dangerous and undesirable products such as deepfakes.
A deepfake is an AI-generated video created to mislead the public or spread fake news and misinformation. If such a video is shared on social media, a significant number of people may not realize that it is fake. So, why do people create deepfakes?
Some reasons for creating deepfakes are the following:
• Blackmailing the subject in the video
• Phishing data
• Spreading misinformation and propaganda, especially on political issues
• Financial fraud
• Used by Ponzi schemes to convince people to join them
• Election manipulation
It’s not hard to spot a deepfake video. Here are some signs that you need to watch out for:
- They lack emotions: If the character in the video does not seem to be sad while talking about a tragic event, that’s obviously a deepfake.
- Eyes movements are not natural: Characters in deepfakes usually don’t move their eyes in a natural way.
- Unnatural body movements: It is pretty easy to spot the difference between natural and made-up movements.
- Unnatural skin tones: AI hasn’t gotten good enough to create skin tones that are 100% identical to people’s natural ones.
- Hair looks plastic: Natural hair needs to move effortlessly, especially if the character is moving their head.
- Audio and video are not synced: Most of the deepfakes’ creators spend more time on the video and forget to fully sync it with the audio. If you notice the lips of the character are not in sync with the words being said, then the chances of that video being a deepfake are high.
The secret to recognise deepfake videos is by looking out for unnatural patterns. If anything seems odd in the video, chances are the video was created using AI. If you suspect a video is deep fake, view it on a larger screen.