Short answer
What to know first
To spot a possible AI image, inspect the source, context, small visual details, and emotional claim together. Odd details can help, but they are not proof. Verification matters more than guessing from pixels.
AI images can look polished, emotional, and believable. They can also be mixed with real photos, edited screenshots, misleading captions, or old images used in a new situation.
Older advice focused on strange hands, warped text, or mismatched reflections. Those clues can still help, but modern tools are improving. Start with the source and story, then inspect the picture.
Practical red flags
- No original photographer, publication, date, location, or official source is provided.
- The image is attached to a shocking claim but only appears as a repost or screenshot.
- Background objects, signs, text, shadows, reflections, jewelry, teeth, ears, or hands look inconsistent.
- Faces look too smooth, repeated, waxy, or oddly lit compared with the scene.
- The image pushes a strong emotional reaction before giving traceable facts.
- A product, charity, investment, or celebrity endorsement appears only in an ad or social post.
What to do next
- Look for the earliest version of the image or claim.
- Search the headline, caption, or key phrase from the post.
- Check official social accounts or websites for the person, agency, school, or business named.
- Use reverse image search when possible, especially for disasters, arrests, missing persons, or viral outrage.
- Do not donate, buy, or share because a picture made you feel rushed.
- If the image targets a real person, avoid spreading it until a trusted source confirms the context.
Start with context before zooming in
Ask who is showing you the image and why. A real photo from an unknown account can still mislead. An AI image from a labeled art account may be harmless. Context changes the risk.
For news, disasters, local crimes, school incidents, and political claims, look for a primary source. That might be a city agency, school district, local newsroom, official statement, court record, or original photographer.
Visual clues that deserve a closer look
AI images sometimes struggle with fine details, repeated patterns, readable text, precise logos, reflections, depth, and physically consistent lighting. But cameras, compression, motion blur, and normal editing can create strange artifacts too.
Use visual oddities as a reason to verify. Avoid using them as a public accusation that something is fake.
- Look for inconsistent shadows or reflections.
- Check readable words on signs, badges, packaging, or documents.
- Compare faces, hands, ears, teeth, and glasses only as part of a bigger check.
- Watch for repeated background people, impossible objects, or distorted architecture.
Screenshots need extra caution
Screenshots are easy to crop, edit, stage, or detach from context. A screenshot of a text conversation, news headline, receipt, school notice, bank alert, or celebrity post should be treated as unverified until you can find the original.
If a screenshot asks for money, login codes, document uploads, or urgent sharing, move to a known official channel before acting.
Trusted resources
These references support the guidance above. They do not create certainty about whether specific content is AI-generated.
- BBB: How to spot a deepfake and avoid scams BBB tips for deepfake audio, video, celebrity scams, and business payment scams.
- CISA: Tactics of Disinformation CISA resource explaining tactics such as synthetic media, impersonation, and manipulated context.
- NIST: Reducing Risks Posed by Synthetic Content NIST overview of transparency, provenance, watermarking, detection, and synthetic content risk reduction.
- News Literacy Project: Teaching About AI Free educator tools for AI literacy and critical thinking.