AI, Algorithms & Adab: Navigating Ethical AI in Islamic Business

Artificial intelligence is not neutral. Every algorithm encodes assumptions, every training dataset reflects historical biases, and every automated decision carries ethical weight. For businesses operating within an Islamic framework, this creates both a challenge and an opportunity: the challenge of applying centuries-old ethical principles to genuinely novel technologies, and the opportunity to contribute a values-grounded perspective to a global conversation that desperately needs one.

The Shariah Lens on Automated Decision-Making

Islamic jurisprudence has always been concerned with the conditions under which decisions affecting people’s lives can be delegated. The classical concept of wilayah (guardianship or oversight) implies that consequential decisions — those affecting livelihood, family, or honour — require accountable human agents, not anonymous processes.

Applied to AI, this creates a meaningful question: at what point does algorithmic automation violate the principle of accountable stewardship? The answer is not a blanket prohibition on automation; Islamic tradition has long accepted that intermediaries and proxies can act on behalf of principals. The issue is accountability and transparency — amanah in its fullest sense.

A credit-scoring algorithm that denies financing without explanation, a hiring model that discriminates on proxies for protected characteristics, a recommendation engine that exploits psychological vulnerabilities to maximise engagement — these all represent, from an Islamic ethical standpoint, failures of amanah: they are systems that exercise power without assuming responsibility.

Data as Amanah

The Islamic concept of trust extends naturally to data stewardship. Muslim scholars are beginning to articulate a framework in which personal data is not merely property to be bought and sold, but a trust held by those who collect it. Misuse of personal data — for manipulative targeting, unauthorised profiling, or sale to third parties without informed consent — becomes not merely a regulatory violation but a moral one.

This framing has practical implications for Islamic financial institutions, halal e-commerce platforms, and Muslim-majority digital markets. Organisations that take data stewardship seriously — implementing genuine consent mechanisms, limiting data collection to necessity, and providing real transparency about how data is used — are not merely complying with GDPR. They are enacting Islamic ethics.

Where the Research Is Going

GIMAC’s 2026 conference features a dedicated track on AI, digital transformation, and Islamic business ethics — reflecting the growing urgency of these questions in the academic community. Key themes include:

Algorithmic fairness in Islamic finance. As banks and takaful operators deploy AI for underwriting and fraud detection, researchers are examining whether these models systematically disadvantage communities with limited credit histories — often the most financially excluded Muslim populations.

Shariah screening in the age of machine learning. Traditional Shariah compliance screening relies on human jurists reviewing financial instruments. AI-assisted screening raises questions about whether the reasoning of a model can be validated against scholarly standards.

The ethics of personalisation. Recommendation engines and personalised advertising can serve consumers — or exploit them. The distinction matters both commercially and ethically.

The conversation between Islamic scholarship and AI ethics is early but promising. What the Islamic tradition brings — a sophisticated vocabulary for discussing trust, accountability, proportionality, and the common good — is precisely what the AI ethics conversation needs more of.