Home Papers Patents Talks Posts Experience Projects Contact
Back to publications

arXiv

Decentralized Deepfake Detection Blockchain Network using Dynamic Algorithm management

Dipankar Sarkar

arXiv preprint arXiv:2311.18545 Cited by 7

Abstract

This paper presents a novel approach to deepfake detection through a decentralized blockchain network that dynamically manages detection algorithms. The system leverages blockchain technology to create a transparent and secure platform where multiple deepfake detection algorithms can be deployed, evaluated, and updated in real-time, ensuring adaptability to new deepfake techniques.

Summary

The rapid advancement of deepfake technology poses significant challenges to digital media authenticity. Our research introduces a blockchain-based solution that:

  • Implements a decentralized network for deepfake detection
  • Provides dynamic algorithm management capabilities
  • Ensures transparent evaluation and updating of detection methods
  • Creates a collaborative environment for researchers and developers

Frequently Asked Questions

What is the Decentralized Deepfake Detection Blockchain Network?

The Decentralized Deepfake Detection Blockchain Network is a novel approach to deepfake detection that uses a blockchain-based network to dynamically manage detection algorithms. The system creates a transparent and secure platform where multiple deepfake detection algorithms can be deployed, evaluated, and updated in real-time, ensuring adaptability to new deepfake techniques. Published November 2023 on arXiv (2311.18545). 7 citations as of 2026.

How does the network stay ahead of new deepfake techniques?

The blockchain-based architecture allows multiple detection algorithms to be deployed, evaluated, and updated in real-time on the network. New algorithms can be added as they are developed; underperforming ones can be deprecated. The blockchain provides a transparent audit trail of which algorithm flagged which content. This is fundamentally different from monolithic deepfake detection systems that ship a single model.

Who is this paper for?

The paper is for: deepfake detection researchers, content authenticity / provenance researchers, blockchain infrastructure teams, social media platforms concerned about content integrity, and security researchers studying media manipulation. It is also relevant to policy makers thinking about content authenticity regulation.

Related Content