Frequently Asked Questions
What does the Image Forensic Scanner detect?
The scanner detects AI-generated images from Midjourney, DALL-E 3, Stable Diffusion, Adobe Firefly, and GAN-based generators. It also identifies deepfake face swaps, photo manipulations (splicing, cloning, inpainting), and checks for C2PA content credentials and EXIF metadata inconsistencies.
What image formats are supported?
Supported formats include JPEG, PNG, WebP, TIFF, and HEIC. Maximum file size is 20MB. For best results, upload the original file rather than a screenshot or recompressed version, as recompression can degrade forensic signal quality.
What is Error Level Analysis (ELA)?
ELA re-saves an image at a known compression quality and compares it against the original. Manipulated regions show different error levels, making splicing, cloning, and object insertions visually detectable.
What are C2PA content credentials?
C2PA is an open standard for cryptographically signing media with provenance metadata — who created it, what tools were used, and whether it was AI-generated. AIGeneratedIt verifies C2PA credentials on every scan.
Can it detect Midjourney and DALL-E images?
Yes. XceptionNet and Hive Moderation API are trained on Midjourney V5/V6, DALL-E 3, Stable Diffusion XL, and Adobe Firefly outputs — achieving 95% detection accuracy.
Is the image forensic scanner free?
Yes. Completely free, no account required. Upload any image and receive instant forensic results including GAN fingerprint analysis, ELA maps, and C2PA credential verification.