Separate the claim from the screenshot, clip or quote carrying it. Find the original source, identify exactly what is being asserted, look for independent confirmation, and downgrade confidence when the evidence is only reposts of the same source.
Start with source and context
The question is not simply whether something looks AI-generated. Start with who published it, where it appeared first, what independent evidence supports it and whether provenance information survives from the original file.
A viral claim can spread through many accounts while still tracing back to one unverified source. That is the anchor for the rest of the check: build confidence from source, context and corroboration rather than trying to force a verdict from one clue.
What matters in practice
A viral claim can spread through many accounts while still tracing back to one unverified source.
Synthetic media quality keeps improving, while genuine media can look strange after compression, cropping or editing. That makes appearance alone a poor basis for a confident verdict.
Screenshots remove context and are easy to edit or misattribute.
Provenance can strengthen a conclusion when it is available, but missing provenance is not proof of fakery. Treat it as one piece of evidence alongside the original source and independent corroboration.
Primary documents, official incident reports and independent reporting carry different evidential weight from commentary or reposts.
The consequence of being wrong should set the verification threshold. A harmless meme can tolerate uncertainty; a claim that could move money, damage a reputation or trigger panic deserves a much stronger evidence trail.
Why visual clues are not enough
Verification rarely produces perfect certainty from one clue. Strong conclusions come from several independent pieces of evidence that point the same way, while missing or conflicting evidence should lower confidence.
- Official sources can be incomplete or wrong, so “official” is not identical to “true.”
- Absence of confirmation is not always proof that a fast-moving claim is false.
What could change the answer?
Confidence should move when a better source appears, provenance becomes available or genuinely independent reporting confirms or contradicts the claim. A good verification process leaves room for the answer to change.
What you can do now
Slow the claim down. Find the original source, look for context and corroboration, and increase the evidence threshold when being wrong would have serious consequences.
- Write the claim in one sentence so you know what you are checking.
- Find the earliest primary source and note exactly what it does and does not say.
- Wait for independent corroboration before sharing consequential claims as fact.
A four-question reality check
When a claim sounds dramatic, work through these four questions before deciding how much weight to give it.
What can the system demonstrably do?
What tools, credentials and networks can it actually reach?
Can it keep operating after errors or intervention?
What real-world outcome follows if safeguards fail?
Bottom line
A good verification process is comfortable with “not enough evidence yet.” That is usually safer than forcing every suspicious image, video or claim into a confident real-or-fake label.
Separate the claim from the screenshot, clip or quote carrying it. Find the original source, identify exactly what is being asserted, look for independent confirmation, and downgrade confidence when the evidence is only reposts of the same source.
Sources & further reading
These links are selected for the claims and methods discussed above. Primary or official sources are preferred where available.
Two useful questions
What is the biggest caveat?
Official sources can be incomplete or wrong, so “official” is not identical to “true.” Absence of confirmation is not always proof that a fast-moving claim is false.
What is the most useful next step?
Write the claim in one sentence so you know what you are checking. Find the earliest primary source and note exactly what it does and does not say.