Beyond AI Text Detection: Dr. Ahmed Zayed's AI Fingerprint as an Integrated Multi-Engine and File-Forensic Framework for Auditable Authorship Evidence

This paper presents AI Fingerprint, an integrated multi-engine and file-forensic framework developed by Dr. Ahmed Zayed to support transparent, reproducible, and auditable assessments of AI-assisted authorship. Unlike conventional AI-text detection tools that rely on the output of a single detection engine, AI Fingerprint combines and compares evidence from multiple AI-generated-text detection engines. The framework preserves agreements, disagreements, confidence variations, and inconclusive results instead of reducing heterogeneous findings to an unsupported binary verdict. The framework also performs file-level forensic examination of supported digital documents. This examination may include available authorship properties, application and producer identifiers, creation and modification timestamps, editing traces, document structure, embedded components, revision-related artefacts, and other technically recoverable indicators from DOCX and PDF files. AI Fingerprint separates four evidentiary layers: text-detection results, file-forensic artefacts, contextual observations, and expert interpretation. The resulting report is designed to preserve the provenance of individual findings and make the analytical process reviewable and auditable. The framework is intended for academic-integrity reviews, document-authenticity assessments, research verification, and other authorship-related investigations. It does not treat an AI-detection score, metadata field, or isolated forensic artefact as conclusive proof of authorship. Instead, it adopts evidence triangulation, explicit uncertainty, reproducible reporting, and human expert review. This paper describes the framework’s conceptual foundation, multi-engine architecture, document-forensic workflow, evidence-classification model, reporting structure, practical applications, limitations, and requirements for responsible interpretation.