Drop a document or click to choose
PDF · DOCX · TXT · MD — text is extracted and analyzed. Scanned/image PDFs are not OCRed.
Drop an image or click to choose
PNG · JPEG · WEBP · TIFF — provenance metadata, spectral & noise forensics, ML classifier. You can also paste (Ctrl/Cmd-V) an image anywhere.
Drop a video or click to choose
MP4 · MOV · WEBM · MKV, up to 100 MB — provenance, frame-by-frame classification, temporal noise forensics. First 90s analyzed; the verdict survives closing the tab.
Submit a specimen.The verdict renders here.
Your label joins the training ledger and improves the detector.
◉ Logged. Thank you — filed for training.
initializing…