Is this photo real?
Verify it in seconds.
Verimint's Media Verify detects AI-generated and manipulated images using C2PA provenance, metadata forensics and compression analysis — and shows you the evidence behind every verdict.
Try Media Verify — free No account needed · JPEG, PNG, WebP, AVIF up to 20 MBFour independent layers of analysis
A single signal can be fooled. Media Verify cross-examines every image with independent forensic layers and merges them into one transparent score.
C2PA provenance
Reads cryptographically signed content credentials (Content Authenticity Initiative). A valid signature declaring AI generation — such as trainedAlgorithmicMedia — is treated as hard evidence.
Metadata forensics
EXIF analysis reveals the real history of a file: camera make and model, editing software like Photoshop, stripped or inconsistent metadata.
Compression artifacts (ELA)
Error Level Analysis exposes regions that were re-compressed, spliced or retouched — invisible to the eye, visible to the algorithm.
AI classifier
A machine-learning layer estimates the probability of synthetic generation. When the classifier is unavailable, the rest of the ensemble keeps working — graceful degradation by design.
Verdicts you can defend
Media Verify never answers with a bare "fake" or "real". You get a category, a confidence level and the full evidence trail.
Upload the image
Drag & drop a photo or paste it from a social media post. Optionally add the post's text for context.
Ensemble scoring
Each layer returns a score, confidence and evidence. A missing signal never raises suspicion — it only lowers confidence.
Transparent verdict
Authentic, AI-generated, manipulated or unverifiable — with per-layer breakdown, flags and the file's SHA-256 fingerprint.
Audit trail
Every verification is logged, so your team can document why a decision was made — essential for newsrooms and compliance.
Built for people who can't afford to be wrong
Newsrooms & fact-checkers
Verify user-generated content before publication. Check provenance of viral images in breaking news situations in seconds.
Public institutions
Counter disinformation campaigns with documented, evidence-based image verification aligned with EU transparency requirements.
Businesses & platforms
Screen visual content for AI manipulation in claims, marketplaces and user reports. Prepare for EU AI Act Article 50 transparency obligations.
Frequently asked questions
How does Media Verify detect AI-generated images?
It combines independent analysis layers: C2PA content credentials, EXIF metadata forensics and Error Level Analysis of compression artifacts. An ensemble model merges the signals into a verdict with a confidence level.
Can a missing C2PA signature prove an image is fake?
No — most authentic photos have no C2PA signature at all. Media Verify never treats a missing manifest as evidence of manipulation; it only lowers the confidence of the verdict.
What image formats are supported?
JPEG, PNG, GIF, WebP and AVIF up to 20 MB. The format is detected from the file's actual bytes, not its extension.
Does it work with screenshots from social media?
Usually such images are stripped of metadata and re-compressed, so the honest verdict will often be "unverifiable" with low confidence — which is itself valuable information for an editorial decision.