Imagine watching a video where a famous person says something shocking—but later you find out, they never said it at all. That video may have been a deepfake.

A deepfake is a fake video, audio, or image created using artificial intelligence (AI). The name comes from combining “deep learning” (a type of AI) and “fake.” These videos look very real, but they are made to trick people.

To make a deepfake, a computer program studies thousands of pictures or videos of a person’s face and voice. Over time, the program learns how the person moves, talks, and sounds. Then, it creates new content—like a video of the person saying or doing something they never actually did.

Some deepfakes are just for fun, like putting a celebrity’s face on a dancing body in a meme. But others are more serious—and dangerous.

Deepfakes can be used to:

• Spread lies or political propaganda
• Fool people into scams
• Create fake news
• Damage someone’s reputation

For example, someone might make a deepfake of a world leader declaring war or a teacher saying something offensive. If people believe the video is real, it could cause panic, anger, or harm to innocent people.

Yes—but it’s getting harder. Experts look for signs like:

• Weird blinking or unnatural eye movement
• Odd lighting or blurry faces
• Unusual voice patterns

There are also AI tools designed to catch deepfakes, but the technology is like a race: as new tools catch deepfakes, deepfake creators find new ways to avoid detection.

• Question what you see online. Just because a video looks real doesn’t mean it is.
• Look for clues—check the source, look for strange movements or glitches.
• Get news from trusted sites. Don’t believe everything you see on social media.

Being smart online means staying alert and asking questions. In a world of deepfakes, the truth matters more than ever.

What does the term “deepfake” mean?

How does a computer create a deepfake?

Why can deepfakes be dangerous?

What are some signs that a video might be a deepfake?

How can you protect yourself from being fooled by a deepfake?