Why are deepfakes becoming so realistic and how are they detected?

Deepfakes are becoming super realistic because they use computer tricks to make fake videos look just like real ones.

Imagine you have a robot friend who can copy how people talk and move. This robot uses photos and videos of someone’s face and learns how their eyes blink, mouth moves, and head turns. Then it puts that on another person's face, like changing the skin of a cartoon character! That’s how deepfakes work.

How Deepfakes Are Made

It’s like giving your robot friend special glasses that let it see every tiny detail in videos. The more videos the robot sees, the better it gets at copying faces and voices. Soon, it can make fake videos so good you might not even notice!

How We Catch Deepfakes

Now imagine a detective dog who sniffs out fakes. This dog looks for clues like odd blinking or unnatural movements. Sometimes, deepfakes don’t move exactly like real people, just like how your robot friend might trip over its own feet sometimes! With the right tools, we can spot these tiny mistakes and know it’s not really that person talking.

Sometimes, even the detective dog needs help from other detective friends to catch the best deepfakes.

Take the quiz →

Examples

  1. A child sees a video of their favorite celebrity talking to them, but it's actually made by a computer.
  2. Someone uses a fake video to trick their friend into giving away money.
  3. An AI creates a realistic video of someone saying something they never said.

Ask a question

See also

Discussion

Recent activity