ArchivAI Revolution: Pioneering the Fusion of Blockchain, AI, and ML for Digital Preservation : Long Paper - iPRES 2025 - Wellington, New Zealand / Te Whanganui-a-Tara, Aotearoa

Abstract

How can archivists cope with managing large backlogs with limited resources? How will they ensure digital authenticity in an era of deepfakes and AI-generated content now and in the future? To tackle this, we developed ArchivAI, a practical framework that combines blockchain verification with AI processing. Our system is built on a Flask-based API with MongoDB Atlas, keeping documents and metadata separate but linked through unique identifiers. Our approach is unique because AI plays a direct role in verification. When the system extracts text, identifies entities, generates tags, and creates summaries, these actions are cryptographically verifiable. Each document is assigned a unique hash stored on a blockchain, ensuring an audit trail that prevents tampering. We moved beyond theory to build a working prototype. When a document enters the system, it is securely stored, processed by AI for better access, and authenticated using blockchain. Our tests show that this method speeds up processing while strengthening authenticity. ArchivAI provides a practical solution for archives to maintain trust, improve accessibility, and work more efficiently.

Details

Creators
Oluseyi Noah Adebanjo; Anuoluwapo Victoria Alabi; Oyeleke Onaolapo Bolaji
Institutions
Date
2025-09-01 00:00:00 +0000
Keywords
digital humanities
Publication Type
paper
License
http://creativecommons.org/licenses/by-sa/4.0/
Download
308622 bytes

View This Publication


This site uses Just the Docs, a documentation theme for Jekyll.