News, in its purest sense, is meant to convey truthful, unbiased, and informative facts about issues affecting the world. While technologies offer new and easy ways, notably through social media, to disseminate information on a large scale, they can also be used as powerful echo chambers for disinformation campaigns.
Disinformation, includes all forms of "false, inaccurate, or misleading information presented and promoted to intentionally cause public harm or for profit" [4] and erodes trust in institutions and media, hampering the ability of citizens to take informed decisions.
Distributed Ledger Technologies (DLT), of which a notable example is blockchain, support decentralized methods for consensus reaching as well as storing and securing transactions and other data with no central intermediaries.
The blockchain, firstly used as public ledger for the Bitcoin cryptocurrency [1] is a quite novel technology, essentially consisting in a replicated database distributed among nodes belonging to different parties. Recently, it has gained a great momentum for its fascinating properties (e.g., distributed consensus, data integrity, non-repudiation).
Although insofar this technology has mainly been used to develop financial applications, the above reasons it has a strong potential for revolutionary innovations that can be beneficial for the society.
The purpose of this research is devising a blockchain-based infrastructure to enable the assessment of news trustworthiness, with the ultimate goal of fighting online disinformation spread.
It offers public endpoint where users can report news resources, that will undergo a combination of automatic preprocessing and human-curated fact-checking, the latter being performed by a network of trusted members belonging to accredited companies and resulting in a trustworthiness score that will be stored indefinitely along with the news metadata, as a transaction of the public ledger managed by a publicly accessible blockchain.
In the following, we elaborate along two directions on the novelty and potential impact of the project:
Blockchain for social good
Only two among the mayor media platform operators have means for providing crowd-powered procedures to fight disinformation spread, in a complex tradeoff between time for getting adequate flagging and exposures. However, flagging options are deliberately hidden to not interfere with the news feed experience. Furthermore, providers have the data to match readers, articles and ads, and are gaining market power and revenue, thous being subject to a clear conflict of interest.
Our proposal offers a viable mechanism to enact this process through a highly innovative technology in a secure and decentralized way, where users submit information to be checked by a clever mixture of automatic and human-curated processes, in a way that preserves the anonymity of the source, the integrity of the submission and also allows for a virtue cycle where good news providers are recognized for their effort.
To the best of our knowledge, we are not aware of any previous approach that attempts to pursue these goals in an unified way.
Moreover, in this specific direction goes the rising interest, both nationally and internationally, towards applications of blockchain technologies for social good, that can "facilitate users' assessment of content through indicators of the trustworthiness of content sources and provide easily accessible tools to report disinformation", and "contribute to a better online space, increasing cybersecurity and trust in online services" [5].
Furthermore, the application will provide an API allowing websites and providers to reference their resources once validated via the blockchain. This fosters virtue cycle where good news providers can be recognized for their effort.
Novel setting:
Concerning our blockchain infrastructure, several peculiar aspects of the problem must be taken into account:
i) The initial, submission phase must guarantee a high level of data integrity and a high degree of anonymity, to limit any form of censorship and manipulation of the reported content (Proof of Work, or analogous is required ).
ii) Conversely, the second, human-operated phase will require the explicit authentication of fact-checkers (i.e. their identity must be publicly disclosed), that will sign each assessment, de-facto securing the whole process (Proof of Authentication) guaranteeing non-repudiation.
iii) The ledger must be publicly accessible at every stage of the pipeline
Such requirements deem necessary adopting a hybrid approach, namely a cascaded, two-layer blockchain where the first layer is public and permissionless, and a second private, permissioned blockchain of fact-checkers, and the two layers are connected using an anchoring mechanism.
The anchoring [20] technique is a timed operation that permits linking a specific block of the first level blockchain with a block of the second.
This context is akin to all such cases where public access, high data integrity as well as explicit authentication, must be jointly ensured, yet, except a few exceptions [22-23] where the inverse, and fundamentally different case is considered, it remains largely unexplored.
Along with issuing scientific publications, we plan on early releasing a white paper documenting our infrastructure, both to foster continuing research on the topic and to gather expressions of interests in the light of obtaining a network of interested parties who can act as our validators, with the aim of deploying a production-level infrastructure which can be efficiently and effectively used in practice.
References:
[11] Nakamoto, S.. "Bitcoin: A peer-to-peer electronic cash system." (2008).
[12] Conroy N et al. Automatic deception detection: Methods for finding fake news. ASIS&T, 2015.
[13] Rubin VL et al. Deception detection for news: Three types of fakes. ASIS&T, 2015.
[14] Dong, XL et al. Knowledge-based trust: Estimating the trustworthiness of web sources. VLDB, 2015.
[15] Ciampaglia GL et al. Computational Fact Checking from Knowledge Networks. PLOS ONE 2015.
[16] Tschiatschek S et al. Fake News Detection in Social Networks via Crowd Signals, Comp. WWW, 2018.
[17] Jooyeon K et al. Leveraging the Crowd to Detect and Reduce the Spread of Fake News and Misinformation. WSDM, 2018.
[18] Tabibian B et al Distilling Information Reliability and Source Trustworthiness from Digital Traces. WWW, 2017.
[19] Narwal V et al. Automated Assistants to Identify and Prompt Action on Visual News Bias. CHI EA, 2017
[20] Pilkington, M. Blockchain technology: principles and applications. Research handbook on digital transformations, 2016.
[21] Huckle S et al. Fake news: a technological approach to proving the origins of content, using blockchains. Big Data, 2017.
[22] Gaetani E et al. Blockchain-based database to ensure data integrity in cloud computing environments. ITA-SEC, 2017.