Artificial intelligence (AI) is already in phones, study, and work. It brings benefit, yet also risks for faith, privacy, and time. Below are brief signposts in the logic of the Code and the Hanafi school.
Short takeaway
- The default is permissibility, within الشَّرِيعَةُ (Sharia) and مَقَاصِدُ ٱلشَّرِيعَةِ (objectives of Sharia).
- Responsibility stays with the human; decision‑making must remain with humans.
- Privacy must be protected, and ḥarām (حَرَامٌ) content must be filtered.
- Religious questions need verification by a trustworthy ‘ālim (scholar).
- Usage must be logged, and limits must be published; transparency improves safety.
Allowed / Caution / Not allowed
Reference: the guidance relies on relevant Code clauses (e.g., 2.2.2, 2.2.4, 2.2.18, 2.2.30, 2.2.33, 2.2.35, 2.2.37, 2.3.8, 4.1–4.3).
| Allowed | Caution | Not allowed |
| Study, translation, note‑taking, time planning. | Sharing personal data and photos with services — keep minimal and deliberate. | Generating, viewing, or spreading ḥarām content. |
| Work help: email drafts, résumés, and code. | Auto‑generated religious answers without review by scholars. | Delegating Sharia rulings to AI in ‘ibādāt and mu‘āmalāt. |
| Picking ḥalāl alternatives, checklists, reminders. | Medical or financial decisions without a human in the loop. | Fraud, deepfakes, and invasion of privacy. |
| Sourcing facts for self‑education. | Long sessions — risk of addiction and procrastination. | Using stolen data or illegal software; violating copyright. |
| Home cyber‑hygiene: filters and parental controls. | Uploading photos of children or relatives — only if clearly necessary. | Manipulation, slander, and spreading unverified claims as “truth”. |
How to act in practice (step‑by‑step)
- Intention (نِيَّةٌ): the goal must be lawful and beneficial.
- Tool choice: a service with a clear data policy must be chosen.
- Privacy: extra data collection must be disabled; share only what is necessary.
- Filters: 18+ and prohibited topics must be blocked; enable parental controls at home.
- Human‑in‑the‑loop: decision‑making must remain with humans; AI is a draft, not the final result.
- Religious answers: information must be verified with a qualified specialist.
- Logs and limits: usage must be logged; known limits and errors must be published.
- Law: legal norms such as the GDPR (General Data Protection Regulation) and local law must be observed.
- Technology: updates and models must be tested for safety and compliance.
Common mistakes and how to avoid them
- Treating AI like a mufti: an LLM (large language model) is not a mufti. A human ‘ālim must review.
- Personal data: uploading passports or medical records risks leaks. Prefer masking and redaction.
- Addiction: long, aimless sessions. Time limits and goals must be set.
- Copyright: copying paid content. Intellectual‑property rights must be respected.
- Lack of transparency: no logs or limits. Risks and boundaries must be described.
When to consult an imam/specialist
- Questions on ‘ibādāt, commercial law, family matters, and inheritance.
- Doubts about permissibility of a scenario or content.
- Medical and financial decisions that affect health or property.
- Cases of harm to honour/reputation or breaches of privacy.
Conclusion: brief counsel and du‘ā
Brief counsel: the goal must be clear; humans remain responsible; harm must be prevented before benefit; privacy and honour must be preserved.
Du‘ā: «ٱللَّهُمَّ ٱهْدِنَا لِلْحَقِّ وَٱرْزُقْنَا تَقْوَى وَعِفَّةً فِي ٱسْتِعْمَالِ ٱلتِّقْنِيَاتِ، وَٱجْعَلْهَا سَبَبًا لِلْخَيْرِ، وَٱصْرِفْ بِهَا عَنَّا كُلَّ شَرٍّ. آمِين.»
Disclaimer: The advice is informational and not a fatwa (فَتْوَى). In doubtful cases, consult a trustworthy ‘ālim (scholar).
