Using Blockchain to Rein in The New Post-Truth World and Check The Spread of Fake News

Adnan Qayyum, Junaid Qadir, Muhammad Umar Janjua, Falak SherarXiv 2019 · arXiv

TL;DR

This paper proposes a decentralized, blockchain-based framework designed to restore trust in the digital news ecosystem. By combining smart contracts for publisher identity management with semantic similarity analysis for content integrity, the authors outline a tamper-proof architecture to detect and mitigate the spread of fake news and deepfakes.

What Problem Does This Paper Address?

We are living in a "post-truth" era where the rapid dissemination of misinformation poses a fundamental threat to democracy and social stability. The authors identify three primary drivers of this crisis:

  1. Digitization and Social Media: The ease of sharing content on platforms like WhatsApp and Facebook allows rumors to go viral instantly.
  2. Big Data and Filter Bubbles: Personalized algorithms isolate users into echo chambers, where confirmation bias makes them more susceptible to fake news that aligns with their existing views.
  3. Generative AI: The rise of Deep Learning (DL) and Generative Adversarial Networks (GANs) has made it possible to create "deepfakes"—eerily realistic but entirely fabricated audio and video.

Current centralized solutions from tech giants like Google and Facebook are struggling to keep up. Furthermore, traditional cryptographic hashing is too "brittle" for news; a single character change or formatting edit results in a completely different hash, making it difficult to verify the integrity of a story as it is reshared and slightly modified across the web.

Methods

The authors propose a high-level blueprint for a "News Blockchain" consisting of three main components:

  • Publisher Management Protocol: This uses smart contracts to manage the identities of news organizations. It handles Enrollment (verifying identities via public keys or third-party APIs), Identity Updates, and Revocation (removing bad actors). It also maintains an Evolvable Reputation Set, where publishers gain or lose credibility based on their history of sharing verified content.
  • Smart Contract for News: When a publisher releases a story, this contract stores the metadata (publisher name, timestamp, etc.) and a signed news string on the blockchain. Crucially, the authors suggest using semantic similarity (via tools like word2vec) rather than simple hashing. This allows the system to recognize that two articles are "the same" even if their wording differs slightly.
  • Building the News Blockchain:
    • Consensus: The system uses Proof-of-Authority (PoA), where a majority of "honest miner nodes"—operated by credible, mainstream media outlets—maintain the ledger.
  • Verification: A Proof-of-Truthfulness (PoT) method uses Merkle trees to allow any participating node to verify if a piece of news exists on the blockchain in $O(\log n)$ time, ensuring the process is efficient for end-users.
Fig. 1: The proposed architecture of the blockchain-based news verification framework has three components: (1) Publisher Management Protocol; (2) Smart Contract for News; and (3) The News Blockchain.

Datasets

Not discussed in the paper. As this is a conceptual framework and blueprint, the authors do not perform empirical testing on specific datasets.

Key Insights and Ideas

  • Semantic Integrity vs. Cryptographic Integrity: The most significant insight is that news verification requires a "fuzzy" match. By proposing semantic similarity, the authors acknowledge that the meaning of the information is what needs to be verified, not just the specific bit-sequence.
  • Decentralized Editorial Control: The framework shifts the "source of truth" away from a single central authority to a distributed network of reputable publishers (the honest miners).
  • Reputation as a Currency: By tying a publisher's ability to reach audiences to an evolvable reputation score, the system creates a technical incentive for accuracy.
  • Efficient Verification: The use of Merkle trees for Proof-of-Truthfulness ensures that checking a story doesn't require downloading the entire blockchain, making it viable for mobile users and social media applications.

Key Results

As this paper presents a high-level architectural overview and theoretical framework, it does not include quantitative results (such as accuracy percentages or latency measurements). The primary result is the design of the multi-layered blockchain architecture itself.

Limitations and Future Work

  • Subjectivity and Bias: The authors note that incorporating consumer feedback into reputation scores introduces the risk of subjective bias and manipulation by malicious actors.
  • Contextual Nuance: While semantic similarity is a step up from hashing, it still struggles with deep contextual nuances. Future work will need to integrate more advanced Natural Language Processing (NLP) to better quantify trustworthiness.
  • Deepfake Privacy: The authors highlight a looming threat where personal photos are scraped to train GANs. They suggest future blockchain solutions could cryptographically protect personal media so that every interaction is traceable.
  • Prototype Development: The current work is a blueprint; the authors aim to develop a functional prototype to test these concepts in a real-world P2P environment.

Summary updated

Abstract

In recent years, `fake news' has become a global issue that raises unprecedented challenges for human society and democracy. This problem has arisen due to the emergence of various concomitant phenomena such as (1) the digitization of human life and the ease of disseminating news through social networking applications (such as Facebook and WhatsApp); (2) the availability of `big data' that allows customization of news feeds and the creation of polarized so-called `filter-bubbles'; and (3) the rapid progress made by generative machine learning (ML) and deep learning (DL) algorithms in creating realistic-looking yet fake digital content (such as text, images, and videos). There is a crucial need to combat the rampant rise of fake news and disinformation. In this paper, we propose a high-level overview of a blockchain-based framework for fake news prevention and highlight the various design issues and consideration of such a blockchain-based framework for tackling fake news.
BibTeX
@misc{qayyum2019using,
  author = {Adnan Qayyum and Junaid Qadir and Muhammad Umar Janjua and Falak Sher},
  title = {{Using Blockchain to Rein in The New Post-Truth World and Check The Spread of Fake News}},
  year = {2019},
  eprint = {1903.11899},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/1903.11899},
}
Figure