New Delhi: At a time when the Indian government is taking its time to frame a comprehensive regulatory framework for cryptocurrencies, researchers in Saudi Arabia are grappling with a whole different crypto issue: is it haram or halal?
What Islamic scholars and the Indian government seem to have in common is concerns around the “intrinsic value” of cryptocurrencies and the gamble inherent in investing in them.
On the Indian front, the Ministry of Finance’s latest Economic Survey, presented to Parliament Tuesday, says crypto assets do not “pass the test of being a financial asset because it has no intrinsic cashflows attached to them”. The topic of cryptocurrencies was also conspicuously absent from Finance Minister Nirmala Sitharaman’s Budget speech Wednesday.
While these are the latest developments or non-developments in India, the government has historically been wary about regulating cryptocurrencies.
Meanwhile, in Islamic countries like Saudi Arabia, a major conundrum is whether or not investing in cryptocurrencies is Sharia-compliant given that it can be akin to gambling, which is haram or proscribed by Islamic law. Other red zones when it comes to Sharia are charging or paying interest and economic speculation.
While various Islamic scholars or clerics have issued fatwas against crypto over the last few years, it has been a theologically grey area in general.
Amidst this uncertainty, an Arab computer science researcher has proposed a machine learning model, based on inputs from Islamic scholars, that can help Muslims decide which cryptocurrencies they can invest in without falling afoul of Sharia tenets.
The November 2022 research paper by Shahad Z. Al-Khalifa, a student of the King Saud University’s College of Computer and Information Sciences is titled ‘CryptoHalal: An Intelligent Decision-System for Identifying Halal and Haram Cryptocurrencies’, and is available on the open-access and non-peer-reviewed platform arXiv.
“From the Islamic finance perspective, cryptocurrencies have raised many issues among Islamic scholars regarding its compatibility with Sharia… َ particularly if it satisfies the Islamic requirement for a currency such as intrinsic value,” the paper says.
“It has become necessary for Islamic finance to adapt to modern changes and assist in providing products and services to Muslims that are compliant with Sharia,” it adds.
The objective of the research, writes Al-Khalifa, was to “investigate and identify features and characteristics that play a part in determining the jurisprudence rulings on cryptocurrencies”. Based on these identified features, an “intelligent system” containing a machine learning model was designed to classify cryptocurrencies as halal (permissible) or haram (forbidden).
Also read: Lucknow to Ludhiana, small-town women are entering crypto world, leaving behind tech bros
The ‘Crypto-Halal’ system
The machine-learning system designed to help Islamic investors make Sharia-compliant decisions is called ‘Crypto-Halal’.
The research paper says the system currently has a dataset of 106 cryptocurrencies, of which 56 have been classified as halal and 50 as haram.
Among the major cryptocurrencies, Bitcoin, Ethereum, and Dogecoin come under the halal category, while Shiba Inu (SHIB) token, Alpha, and PancakeSwap (CAKE) are labelled haram.
To use CryptoHalal, a user must enter the name of the cryptocurrency. At the back-end the system first checks for a match in the database of cryptocurrencies the researcher has already classified as haram/halal. If a match does not exist, the system queries an external website of CoinMarketCap, a leading website that keeps track of new and existing cryptocurrencies.
Then based on the properties and features of the cryptocurrency, the system will let the user know if the currency is “probably halal” or “probably haram”.
The paper notes that it is important to have an intelligent system that can distinguish between haram and halal crypto since “more than 30 new cryptocurrencies” crop up every month.
Further, the researcher claims that in a Twitter survey conducted among “Muslim cryptocurrency investors”, a need for such a system had emerged.
The survey received 748 responses in all, the paper says. Around 96 per cent of the respondents reported they had cryptocurrencies, and nearly half of these claimed to own more than five.
One of the survey questions was: “Do you think that having an intelligent platform that helps extract the jurisprudence ruling of cryptocurrencies would impact your decision to buy cryptocurrencies?” To this question, 620 responses were received, of which 83.2 per cent agreed that it would indeed impact their decision.
“Our questionnaire shows that such a system like ours is needed in the Muslim community and would solve a large-scale conflict in today’s world,” the paper says.
How does system tell halal from haram?
The system has been trained to decide whether a cryptocurrency is haram or halal based on 20 features commonly found in cryptocurrencies. This classification is based on the interpretations and rulings around cryptocurrencies found on a Telegram channel that is “supervised by Islamic scholars” and named ‘CRYPTOHALAL’ as well.
The Telegram channel specialises in observations and legal audits of the crypto market to “assist Muslims with the jurisprudence ruling on cryptocurrencies”.
“To date, there are no published datasets that contain the classification of cryptocurrencies as Halal or Haram,” the paper says and admits this is an area where the decision-making system can be improved by switching from manually selecting features from the Telegram channel’s rulings to a system that automates this picking and choosing via Deep Learning models.
The system has separated the 20 features commonly found among cryptocurrency projects into two categories called ‘High Priority’ and ‘Low Priority’ based on the Telegram channel run by Islamic scholars.
“High Priority indicates features that have only Haram cryptocurrencies, and Low Priority indicates features consisting of both Halal and Haram cryptocurrencies,” the paper says.
“This separation will assist in determining the cryptocurrency’s degree of compliance with Sharia. For example, if the majority of the high-priority features in the cryptocurrency are satisfied, then it is more likely to be Haram with an outcome of ‘Probably Haram’, otherwise it is more likely to be Halal with an outcome of ‘Probably Halal’,” it adds.
The haram-linked High Priority features include speculation (when a cryptocurrency project is considered to be based on speculation), and borrowing (when a cryptocurrency project provides borrowing services).
The not-so haram Low Priority features include staking (when a crypto project provides some incentive to those who are long-term holders of its cryptocurrency), and governance (where a token of the cryptocurrency project can be used to vote on the direction of the project).
“[W]e have found that 45 out of 50 Haram cryptocurrencies’ projects use Decentralized Finance (DeFi) and liquidity pools, whereas none of the Halal cryptocurrencies participate in lending & borrowing services, leverage, margin, and prediction market,” the paper says.
(Edited by Asavari Singh)
Also read: Binance’s books are a black box, filings show, as crypto giant tries to rally confidence