When Algorithms Decide Who Deserves Sharia Financing

In the era of conventional banking, the decision to grant financing was made by a human credit analyst who examined financial reports, interviewed prospective clients, and assessed character. Today, that process is increasingly being handed over to artificial intelligence (AI) and machine learning. These systems can process millions of data points in seconds, evaluating everything from transaction history and digital wallet usage to social media activity to produce a credit score. While this digital transformation offers remarkable speed and efficiency, it raises a fundamental question for the Islamic finance industry: can an algorithm truly deliver justice?

This question is particularly pressing as Indonesia’s sharia banking sector continues its robust growth. By March 2026, sharia banking assets had surpassed Rp1,061 trillion, with financing growing to over Rp716 trillion, outpacing the national average. The peer-to-peer lending sector is also booming, with outstanding loans reaching Rp98.54 trillion. This rapid digitisation promises to expand access to financial services, but it also carries the risk of creating new, invisible barriers.

The problem lies in the nature of data. AI systems learn from historical data, and if that data reflects existing inequalities, the algorithm will not only inherit but potentially amplify them. A small village trader with a healthy business and steady cash flow might be deemed a higher risk than an urban office worker simply because the trader lacks a rich digital footprint. The system favours those who are already deeply integrated into the digital economy, leaving behind the very groups that financial inclusion agendas were designed to help. This creates a new form of exclusion, not based on a lack of collateral or a bank account, but on a lack of sufficient digital data.

For Islamic finance, this is a critical issue. Discussions on sharia compliance often stop at the contract level, ensuring products use murabahah, musyarakah, or ijarah structures. However, the substance of sharia is much broader, rooted in principles of justice (’adl), public welfare (maslahah), and the avoidance of oppression. If an algorithm systematically blocks certain groups from accessing financing without transparent justification, it contradicts the very objectives of sharia (maqashid sharia), which aim to protect wealth, reduce inequality, and promote productive economic activity. The industry must therefore ensure that the efficiency of AI does not come at the cost of the fairness and inclusivity that define Islamic finance.