A photo arrives with a startling caption. Your first question might be, ‘Is this AI?’ A more useful first question is, ‘Who says this happened, and how would they know?’ A real photo can be cropped, old, or attached to a false story. An AI-made image can look perfectly ordinary.
Start with the source
Look for the original post, photographer, or organization named in the claim. If a major event supposedly happened, see whether reliable local reporting or an official source confirms it. A repost with no place, date, or source gives you very little to work with.
Search for the image
A reverse image search can show where similar pictures appeared earlier. That may reveal an older photo being presented as new, a cropped version, or an original creator. Search results are clues, not a final verdict; new images may not appear anywhere else yet.
Odd text, mismatched reflections, or impossible details can raise a question. But there is no dependable checklist of visual mistakes. AI-made images can avoid obvious errors, and real photos can have strange lighting or blur.
Share with care
If a picture makes you angry or frightened, wait before forwarding it. You can ask the sender where it came from or share a reliable report instead of the unverified image. The goal is not to become a forensic expert. It is to keep one uncertain picture from becoming a false certainty.