A few years ago, a photo or video was usually good enough proof that something happened. Today, software can create a realistic picture of an event that never took place, copy a person's voice from a short clip, or swap one face onto another in a video. These fakes are called deepfakes, and they are getting harder to notice. You do not need to be an expert to protect yourself. You need a handful of habits and a healthy amount of curiosity.
What is a deepfake?
A deepfake is media, such as an image, audio clip, or video, that has been created or altered by AI to look or sound real. The word comes from deep learning, the kind of machine learning used to build these tools, plus the word fake. Not every AI-made image is a deepfake in the harmful sense. A cartoon-style picture made for fun is just a creation. The concern is realistic media used to mislead people, such as a fake video of a public figure saying something they never said, or a cloned voice used to trick someone's family.
Why they matter to students
Fakes spread fast on social media, where people share before they think. They can be used to spread false news, to embarrass classmates, to scam families, or to damage someone's reputation. Creating and sharing fake images of real people, especially without their consent, can seriously hurt them, and in many places it can also break school rules or the law. So there are two skills to build: spotting fakes, and refusing to make or spread them.
Visual clues in images and video
Older fakes had obvious flaws, and some still do. Look closely for these signs, but remember that none of them is proof on its own.
Hands and fingers. Extra fingers, strange joints, or blurry hands used to be a giveaway, though newer tools are improving.
Text and logos. Signs, shirts, and labels in the background may contain garbled or nonsense lettering.
Lighting and shadows. Shadows that point the wrong way or a face lit differently from the scene can signal editing.
Edges and hair. Look for a fuzzy outline around a head, or hair that seems to melt into the background.
Skin and eyes. Skin that looks too smooth, mismatched reflections in the eyes, or unusual blinking can be clues.
Backgrounds. Warped objects, repeated patterns, or crowds with oddly similar faces can show up in generated scenes.
Clues in audio
Cloned voices can sound surprisingly natural. Listen for flat emotion, odd pacing, unnatural breathing, or words that sound slightly metallic. More importantly, think about the situation. If you receive a voice message from a relative asking for money urgently, hang up and call them back on a number you already know. A real person will not mind you checking.
Skip the pixel hunting: check the context
Because the visual clues keep disappearing, the most reliable skills are about context, not pixels.
Who posted it first? Look for the original source. Be suspicious of a dramatic clip with no clear origin.
Did reliable outlets report it? A world-changing event will be covered by multiple trustworthy news organizations, not only one anonymous account.
Does it match what is known? If a clip shows a person in a place or saying something that contradicts the record, pause.
Is it designed to make you angry or afraid? Content built to trigger a strong emotional reaction is more likely to be manipulative.
Handy tools and their limits
A reverse image search can show where else a picture has appeared and whether it is older than the claim attached to it. Some platforms add labels or hidden markers to AI-generated content, and some companies offer detection tools. These can help, but they are not perfect. Detectors can miss fakes and can also flag real images by mistake. Treat them as one clue among several, not as a final verdict.
What to do when you suspect a fake
Do not share it. Even sharing to say "can you believe this?" helps it spread.
Check a trusted source. Look for the same story from reputable news outlets or the person involved.
Report it. Most platforms have a way to flag misleading or harmful content.
Tell a trusted adult if the fake targets you or someone you know. You are not at fault, and help is available.
Making your own responsibly
AI image and video tools are fun and creative, and there is nothing wrong with using them. The line is simple. Do not use them to imitate a real person without their permission, to mislead people into believing something false, or to embarrass or harass anyone. Label your creations as AI-made when you share them. That small habit keeps trust intact.
The mindset that works
You do not need to distrust everything. You need to slow down. Before you believe or share a striking image, clip, or voice message, take ten seconds to ask where it came from and whether anyone reliable confirms it. That pause is the single most effective defense against deepfakes, and it works no matter how good the technology gets.
Keep learning: explore our blog or take the free Introduction to AI course.
Sources: media literacy and AI safety guidance from news literacy organizations and major technology companies.

