Fiqh of Transactions Part 10

Use of AI (Artificial Intelligence) Technology: Initial Requirements under an Ethical Framework

In the title it is called Initial because this topic, through further research, the joint efforts of various governments, scientists, lawmakers, and experts in ethics, needs the development of a practical and implementable system of control (governance/oversight).

(This discussion is not limited to whether something is just halal or haram, but rather focuses on the demands of forming a governance framework related to the use of AI, its political implications, regulations, and rules.)

In today’s era, the use of AI (Artificial Intelligence) technology is rapidly increasing across various sectors.

This technology is not limited to ChatGPT, where you seek information on a certain topic; rather, its scope is much broader.

My learning background regarding this....

During the studies of Master of Business Administration in Information Technology (MBA IT), I had a great opportunity to learn from Professor Dr. Tarique about the Artificial Neural Network (ANN) model and machine learning.

In reality, the way AI technology functions, one of its models is ANN (Artificial Neural Network), which, based on the structure of the human brain, works by interconnecting various neurons, nodes, and units.

ANN (Artificial Neural Network) is inspired and derived from the real biological brain system.

And for this discussion, a detailed article is required.

The usage of AI (Artificial Intelligence) technology:

The practical use of AI is happening in the following ways across different sectors:

• Medical Sector: AI models are used for diagnosis of diseases.

• Financial Sector - Fintech: Decisions on the approval or rejection of loans and other Fintech processess function and are made on the basis of AI.

• Education and Judiciary: AI plays a role in granting access to facilities or decision-making for students and citizens.

• Politics, Culture & Ideology: AI has a profound impact in shaping the ideologies, cultural trends, and political processes.

• Replacing Human Decisions: AI makes decisions in place of humans, and that too through codes, which are often invisible and complex.

The need for a Governance Framework (Structure of Control):

To guide AI technology in the right direction and to protect against its negative impacts, it is extremely necessary to develop a strong, comprehensive, and transparent governance framework.

This framework should include the following key points:

• Ethical Regulations

• Policy Formation

• Oversight & Transparency

• Safeguarding Human Rights & Justice

As a detail and explanation of this governance system’s structure, I will quote an excerpt from that paper by Mufti Faraz Adam (expert in Islamic economics and financial technology) titled “A Shariah Governance on Artificial Intelligence”, which is the subject of his *PhD research at IIUM (International Islamic University of Malaysia), and which, according to him, will be completed in early 2026 and will be publicly available to benefit all.

Jazahullahu khayran wa ahsan al-jazaa.

What is in that Governance Framework?

This Shariah-guided AI governance model draws upon revelation, and industry knowledge & practice to offer a faith-rooted answer to the AI containment challenge.

It covers:

✅ Infrastructure fragility — AI systems rely on massive compute and power grids; outages, cyberattacks, or hardware bottlenecks can paralyse essential services.

✅ Algorithmic opacity — Models can hallucinate, deceive, or produce biased outputs with no traceable logic, making them unreliable in high-stakes domains.

✅ Autonomy escalation — Future AI may self-improve or act beyond intended scope, posing control and alignment risks that current governance cannot contain.

✅ Human liability gaps — When AI harms — through misdiagnosis, misinformation, or automation errors — it remains unclear who is morally or legally accountable. What about the actions of 'humanoids'? When an AI causes harm, who bears responsibility: the user, the developer, or the deployer?

✅ Validity of consent when engaging with opaque systems — e.g., users relying on AI-generated legal or health advice without full understanding of model limitations. What constitutes meaningful and valid consent?

✅ Trust and custodianship (amānah) in handling model outputs and system design — especially when models are deployed in religious education, fatwa-generation, or Qur’an apps. How do we ensure that AI systems don’t distort human values or mislead the public?

✅ Fairness and bias in training data — preventing unjust outcomes in systems trained on culturally narrow or non-representative datasets.

✅ Limits of delegation (Tawkeel-Tafweed) — determining what types of decision-making may be ethically handed over to machines. When does reliance on AI cross into impermissible abdication of human judgment?

✅ Prevention of harm (ḍarar-ضرر) — addressing misuse of AI in manipulating opinions, generating false religious claims, or automating jobs without ethical transition planning.

This isn’t simply about “Halal AI” or restricting use cases — it’s about building a governance ethic rooted in purpose, responsibility, and accountability.

AI is not just another tool. It now:

⚠️ Makes medical diagnoses

⚠️ Approves or rejects loans

⚠️ Determines access to education or justice

⚠️ Shapes culture, politics, and perception

⚠️ Replaces human judgment with code — often invisibly

With generative AI accelerating rapidly, the ethical risks are multiplying!

As an illustration, today's Generative AI is capable of producing poetry, stories, images, even religious fatwas." its ethical risks are also increasing!

It must be remembered that these demands and requirments of Governance Framework are not limited to Islamic principles of halal and haram,

but go beyond, requiring a clear policy and framework for societies, states, and institutions at the political, social, cultural, and legal levels.