Network activity: Runs in your browser. The PDF itself stays in your browser unless explicitly uploaded.
Scan a PDF for fakes
Drop a PDF to check it for metadata anomalies, suspicious producer fingerprints, and signs of AI-generated content. The file stays in your browser — no upload, no account.
What it checks
- /Producer fingerprint — flags known AI / LLM tool strings (ChatPDF, GPT-named producers) + online re-processing tools (iLovePDF, Smallpdf, PDF24) that rewrite signatures and layout.
- Creation + modification dates — missing creation date or massive drift between creation + modification timestamps. Worth a closer look.
- Document structure — page count, AcroForm fields, embedded fonts. Single-page documents with no fields and no images are the AI-gen pattern.
Honest framing: these are heuristics, not cryptographic proofs. A motivated forger can spoof every field below by re-saving through Adobe Acrobat. The scan catches lazy fakes (AI-generated lease agreements, photoshopped pay stubs, recycled bank statements) and gives you a structured report to make a human judgment call. For verifiable authenticity, the source needs to sign with hardware-key sign + you verify the signature.