Media literacy programs increasingly teach students and the public to question whether an image, voice recording, or video is authentic. That is necessary, but it is no longer sufficient.
Synthetic media is becoming easier to create, edit, and distribute. Images can be generated from a sentence. Still photographs can be animated. Voices can be reproduced. Videos can be altered or created without a traditional camera.
In response, many educational programs focus heavily on detection. Learners are shown common visual errors, unusual shadows, inconsistent reflections, distorted hands, unnatural speech, or missing source information.
Detection skills matter. However, a stronger approach also gives learners controlled, practical experience creating synthetic media themselves.
When people understand how generated media is made, they become better at evaluating what these systems can do, where they fail, and why a convincing output should not automatically be treated as evidence.
Detection Alone Creates False Confidence
A checklist of visible errors can be useful, but it can also become outdated quickly.
Learners may be told to look for:
- unusual fingers
- distorted facial features
- incorrect text
- unnatural blinking
- inconsistent lighting
- poor lip synchronization
- strange background details
These signs can indicate synthetic media, but their absence does not prove authenticity.
As generation systems improve, familiar errors become less common. At the same time, authentic images and videos may contain compression artifacts, motion blur, editing, poor lighting, or other imperfections that resemble generated content.
A person who relies only on visual detection may become overconfident.
The better question is not simply, “Does this look fake?”
It is:
- Where did this media come from?
- Is the original source available?
- Has the file been edited or reposted?
- Can the claim be verified independently?
- Is the content being presented in the correct context?
- Who benefits from the audience believing it?
Media literacy should emphasize verification, provenance, and context rather than promising that every synthetic artifact can be identified by appearance alone.
Creation Reveals How the Systems Actually Work
A learner understands generative media differently after attempting to create it.
When students use an image or video generation tool, they quickly discover that the relationship between a prompt and an output is not straightforward.
The same prompt can produce several different results. Small wording changes may alter composition, camera angle, clothing, lighting, or mood. A reference image may guide one part of the output while other details change unexpectedly.
Through practical creation, learners can observe:
- how prompts influence results
- how models interpret ambiguous instructions
- how reference images affect identity and composition
- where visual consistency breaks down
- how editing can hide obvious generation errors
- how several imperfect outputs can be combined into one convincing asset
- how quickly synthetic media can be produced
This knowledge makes synthetic media less mysterious.
A learner who has generated an apparently realistic image understands that realism does not confirm that an event occurred. A student who has animated a still photograph understands how motion can be created without a recorded scene.
The lesson becomes experiential rather than theoretical.
Libraries Are Well Positioned to Teach This
Libraries already help communities evaluate sources, access technology, understand information systems, and navigate new forms of media.
That makes them natural environments for practical synthetic media education.
A library does not need an advanced production studio to run a useful workshop. A simple session can use accessible creative tools, shared example prompts, and a structured discussion.
A basic workshop might include:
- generating an image from a short description
- repeating the generation to compare different outputs
- changing one part of the prompt
- adding a reference image
- animating a still image
- reviewing errors and inconsistencies
- discussing how the output could be misrepresented
- identifying appropriate disclosure language
- verifying a related real-world claim through trusted sources
The objective is not to train professional AI creators.
It is to help participants understand that synthetic media is produced through a sequence of choices, model behaviors, revisions, and editing decisions.
Platforms such as Cliprise can support this type of learning by giving users access to multiple image and video workflows in one environment. The educational value comes from comparing outputs, observing model differences, and discussing why one result appears more credible than another.
Teach the Difference Between Generation and Evidence
One of the most important concepts in synthetic media literacy is that generated media is not evidence of a real event.
This may appear obvious in a classroom exercise, but the distinction becomes less visible when synthetic media is encountered in a social feed, news discussion, private message, or political argument.
A realistic image can illustrate a possibility without documenting reality.
A generated voice can reproduce a style without confirming that a person made a statement.
An animated photograph can create movement without showing something that actually happened.
Learners should be taught to separate several questions:
- Is the media authentic?
- Is the event authentic?
- Is the description accurate?
- Is the source trustworthy?
- Is the media being used as illustration, satire, advertising, or evidence?
- Was the use of AI disclosed?
These questions are related, but they are not identical.
Even authentic media can be misleading when cropped, edited, mislabeled, or removed from context. Synthetic media education should therefore fit within broader information literacy rather than being treated as an isolated technical problem.
Include Consent and Identity in the Lesson
Practical creation also raises questions about consent.
A workshop participant may want to upload a photograph of a friend, family member, public figure, or child. That creates an opportunity to discuss whether technical ability is the same as permission.
Before using a real person's image or voice, learners should consider:
- Did the person consent?
- Is the intended use respectful?
- Could the output damage their reputation?
- Could viewers mistake it for a real recording?
- Does the platform permit this use?
- Should the output include a disclosure?
- Is the person a minor or otherwise vulnerable?
- These issues should not be treated as an afterthought.
They are part of media literacy because synthetic media changes how identity can be copied, altered, and redistributed.
A useful classroom or library policy is to use fictional characters, licensed materials, participant-owned media, or clearly authorized examples whenever possible.
Disclosure Should Be Practical and Visible
Learners should also practice labeling synthetic media.
A vague note hidden in a caption may not be enough when the content could reasonably be mistaken for an authentic recording.
Clear disclosures may include:
- AI-generated image
- AI-generated illustration
- synthetic voice used
- image created from a text prompt
- photograph animated using generative AI
- fictional scene created for educational purposes
The wording should match the medium and the risk of misunderstanding.
Disclosure does not solve every problem, but it helps establish responsible norms. It also teaches learners that authorship includes communicating how content was produced.
A Simple Workshop Framework
Libraries and schools can structure a practical media literacy session around four stages.
Create
Participants generate a basic image or short video from a prompt.
Compare
They produce several variations and identify differences, errors, and unexpected choices made by the model.
Contextualize
They discuss how the media could be presented honestly, misleadingly, or maliciously.
Verify
They practice checking sources, searching for original context, and distinguishing generated illustration from documentary evidence.
This structure keeps the activity focused on critical thinking rather than novelty.
The creative step attracts attention. The comparison and verification steps produce the deeper educational value.
The Goal Is Informed Skepticism, Not Distrust of Everything
Poor synthetic media education can push learners toward one of two extremes.
The first is unquestioning acceptance: if media looks realistic, it must be real.
The second is total cynicism: any inconvenient image, recording, or video can be dismissed as fake.
Both outcomes are dangerous.
The purpose of media literacy is not to make people believe nothing. It is to help them evaluate claims using evidence, sources, context, and appropriate uncertainty.
Hands-on creation supports that goal because it replaces abstract fear with practical understanding.
Learners see that generative systems are powerful, inconsistent, editable, and dependent on human choices. They also see that detection is only one part of verification.
Synthetic media literacy should therefore teach people to create responsibly, disclose clearly, question context, protect consent, and verify independently.
When learners understand how synthetic media is made, they are better prepared to judge how it should be interpreted.
About the Author
Kruno Sulic is the founder of Cliprise, a multi-model platform for AI image and video creation. He works on digital products, creative workflows, model integration, automation, and the practical challenges of making generative media tools understandable and useful.
Connect with Kruno Sulic on LinkedIn.
Enjoyed this article? Subscribe to Video Librarian today for access to over 40,000 pages of film resources tailored for librarians, educators, and non-theatrical audiences.