The best way to help teenagers understand AI-edited video is not to give them a long lecture. Let them compare an original image or clip with a deliberately altered version, then ask them to identify what stayed accurate, what changed, and what cannot be verified.

This 30-minute family exercise teaches observation, consent, and responsible sharing. It does not require posting anything online.

Choose Privacy-Safe Source Material

Start with material your family created and has permission to edit. A drawing, toy, craft project, pet photo, or consenting adult is safer than an identifiable child. Avoid school uniforms, house numbers, private documents, location clues, or images belonging to someone who has not agreed to the activity.

Save the untouched original in a separate folder. Label it “original” so it cannot be confused with later versions.

Make One Controlled Change

Create only one obvious transformation. For example, use an AI photo animation generator to add gentle movement to a drawing or object. Keep the prompt simple: “The paper rocket tilts slightly while the stars shimmer; the camera remains still.”

Alternatively, an adult could use VideoSwap to explore an authorized character transformation, motion-driven animation, or face change involving a consenting adult. Do not use a classmate, celebrity, teacher, or relative who has not provided permission.

The purpose is not to make the most convincing result. It is to create a version that can be examined against a known source.

Compare the Versions in Three Passes

First, watch at normal speed and write down your immediate impression. Did the clip feel believable? Which detail attracted attention first?

Next, pause on several frames. Compare:

  1. Face or character identity
  2. Hands, edges, and small accessories
  3. Text, logos, and repeated patterns
  4. Shadows, reflections, and background objects
  5. Whether motion follows believable physical timing

Finally, return to the original. List details the generated version added without evidence. The unseen side of an object, a new facial expression, or movement between two captured moments is a prediction, not recovered history.

Separate Observation From Assumption

Make two columns titled “What We Can See” and “What We Think Happened.” A statement such as “the person turns toward the window” describes the generated clip. Saying “this is what happened after the photograph was taken” is an unsupported conclusion.

This distinction is central to media literacy. A realistic-looking video may still combine authentic source material with invented motion, expressions, lighting, or context.

Visual clues alone cannot prove that a clip is genuine or altered. Compression, filters, and ordinary editing can resemble AI artifacts, while a polished synthetic clip may contain few visible mistakes. Source, context, disclosure, and corroborating evidence matter alongside appearance.

Discuss Consent Before Sharing

Ask three questions before anyone sends or posts the result:

  1. Did everyone shown agree to this specific transformation?
  2. Could a viewer mistake the edit for a real event or statement?
  3. Does the frame reveal private information beyond the intended subject?

If the answer is uncertain, keep the exercise private. When sharing an altered clip, use a clear label such as “AI-animated from a still image” or “fictional character transformation.” A playful edit should not depend on misleading the audience.

Finish With a Family Rule

End by writing one rule everyone can remember. For example: “Pause before sharing, find the original source, and separate what the video shows from what the caption claims.”

The goal is not to make young people suspicious of every image. It is to teach a better habit: slow down, inspect evidence, respect the people represented, and ask what is known. That skill is useful whether a video was created by AI, edited conventionally, or filmed exactly as presented.