How to verify AI-generated images, video, text and viral claims using provenance, source checks and repeatable workflows rather than detector hype.
The useful question is rarely “does this look AI-generated?” Modern synthetic media can be convincing, while real media can look strange after compression or editing. Verification works better when you check provenance, original context, independent reporting and the claim around the media instead of trusting one visual clue or one detector score.
Walk through source, context, provenance and corroboration checks before trusting a suspicious image, video or viral AI claim.
Pick the question closest to yours. Each guide gives you the answer first, then the evidence, the important caveats and practical next steps.
VERIFY BEFORE YOU BELIEVEDo not rely on fingers, teeth or a single detector score. Start with the source and claim, check where the image appeared before, inspect available provenance or Content Credentials, and treat a missing signal as inconclusive rather than proof that the image is real.
Read guide →
VERIFY BEFORE YOU BELIEVEVerification should focus on provenance, original source, independent corroboration and the request attached to the video. Lip-sync glitches and odd blinking may be clues, but convincing synthetic video can survive a casual visual inspection.
Read guide →
VERIFY BEFORE YOU BELIEVEAI text detectors can sometimes identify statistical patterns associated with generated writing, but their scores are not proof of authorship. False positives and false negatives matter, especially when a score is used to accuse a student, employee or writer.
Read guide →
VERIFY BEFORE YOU BELIEVEContent Credentials are a standards-based way to attach tamper-evident provenance information to digital media. They can help show how a file was created or edited, but they do not turn the file into guaranteed truth.
Read guide →
VERIFY BEFORE YOU BELIEVEReverse image search is most useful for finding context: earlier appearances, different captions, source pages and visually similar versions. It cannot prove that a never-before-seen image is genuine, but it can quickly expose recycled or miscaptioned media.
Read guide →
VERIFY BEFORE YOU BELIEVESeparate 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.
Read guide →We label controlled evaluations, real-world incidents, expert forecasts and speculation differently. Read the methodology before treating any probability or timeline as a fact.