Technology as a Tool and a Trust

1. Divine knowledge has no technological competitor

Qur’an 2:255 attributes complete life, sustaining authority and encompassing knowledge to اللّٰه (ALLAH). The phrase about created beings encompassing nothing of His knowledge except what He wills prevents a Muslim from speaking of any machine as all-knowing.

Qur’an 96:5 says that اللّٰه (ALLAH) taught humanity what it did not know. Engineers design models and people gather data, but their intelligence, time, materials, discoveries and opportunities are created gifts. Tawheed traces means back to the true source without denying the human work carried out through those means.

Each gold Qur’an panel uses the locked Uthmani Arabic catalogue and separately named English translations. The page keeps six categories distinct: direct Qur’anic teaching, authentic hadith, qualified fiqh application, current technical evidence, current law or regulation, and Tawheed Club’s educational guidance. Technology changes quickly; a current technical description is not revelation, and a general fiqh statement is not an automatic verdict for every tool or use.

Tawheed boundary: اللّٰه (ALLAH) alone is the Creator, the Knower of the unseen and the source from whom all created ability and permitted knowledge ultimately come. People build machines from created matter and use faculties they did not originate. An AI system is a tool: it is not divine, prophetic, a rival creator, a moral authority or a source of revelation. Fluent output is not proof of consciousness, wisdom, truth or access to the unseen.

Words such as “oracle”, “omniscient” or “god-like” are not harmless when they confuse technical capability with divine attributes. Describe the actual function: prediction, classification, retrieval, generation or automation. Precise language protects belief and improves technical judgement.

For adults teaching children: demonstrate with fictional prompts and non-animate sample media. Teach children never to enter a real name, face, voice, school, address, location, medical detail or private conversation into an AI tool. Keep adult-controlled accounts and privacy settings, supervise generated content, explain that convincing output can be wrong, and provide a simple route for reporting anything frightening, sexual, hateful, manipulative or secretive.

2. AI has no independent access to the unseen

Qur’an 31:34 locates knowledge of the Hour, rain, what is in wombs, what a soul will earn tomorrow and where it will die with اللّٰه (ALLAH). A model can estimate from available data; an estimate is not knowledge of the unseen and may fail.

Qur’an 6:59 declares that the keys of the unseen are with اللّٰه (ALLAH) and describes His exhaustive knowledge of creation. A generated prediction about destiny, divine acceptance, a person’s hidden intention or a future event must never be presented as revealed certainty.

Forecasting and unseen knowledge are different categories. A weather model uses measurements and probabilities. A language model predicts sequences from learned patterns. Neither becomes a prophet or receives revelation. Even a useful prediction needs a stated timeframe, evidence and uncertainty.

Do not ask a chatbot to tell whether a deceased person is saved, whether a dream guarantees an event, whether a prayer was accepted or what اللّٰه (ALLAH) has decreed for a named person. Those prompts invite fabrication about matters beyond the tool and beyond ordinary human access.

Evidence practice: before using or sharing a digital output, write who or what produced it, the original source, the date, the intended use, the people affected, the personal data involved and the consequence if it is wrong. Check the claim against a named primary or authoritative source, inspect the surrounding context, look for independent confirmation and record uncertainty. For a religious claim, locate the complete Qur’an passage or authentic report and ask a qualified human scholar about its application. Do not cite an AI system as though it were the source of sacred knowledge.

3. Permissibility in principle is not permission for every use

International Islamic Fiqh Academy Resolution 258 (3/26): Artificial Intelligence says the default ruling for developing and using AI is permissibility, while attaching conditions: lawful purpose and outcome, benefit, prevention of harm, respect for beliefs and symbols, protection of information and rights, security, honesty, documentation and transparency.

The conditions matter as much as the headline. A translation aid and a voice-cloning fraud both use AI; their purpose, consent and harm differ. A study helper and an exam-cheating service are not one ethical act. A private brainstorming session and publication under a false human byline raise different duties.

Ask five questions before use: Is the task lawful? What benefit is expected? Who could be harmed? Which data or rights are involved? What human review and disclosure are necessary? If the answers are vague, the correct next step is not a confident prompt but a pause.

Local law, employer policy, school rules, platform terms and intellectual-property duties may add obligations. The institutional resolution does not cancel those sources or provide a licence to copy protected material.

Evidence practice: before using or sharing a digital output, write who or what produced it, the original source, the date, the intended use, the people affected, the personal data involved and the consequence if it is wrong. Check the claim against a named primary or authoritative source, inspect the surrounding context, look for independent confirmation and record uncertainty. For a religious claim, locate the complete Qur’an passage or authentic report and ask a qualified human scholar about its application. Do not cite an AI system as though it were the source of sacred knowledge.

4. Responsible use of tools cannot be delegated

Qur’an 17:36 makes hearing, sight and heart answerable and forbids following what lacks knowledge. A user who publishes a generated accusation remains responsible for choosing to publish it.

NIST Generative AI Profile 600-1 warns that generative systems can confabulate and can create risks involving information integrity, privacy, security and human reliance. A system may provide a polished explanation for a false answer or invent citations that look credible. Confidence of tone is not a reliability score.

Human review means more than glancing at the output. The reviewer needs enough subject knowledge to detect likely error, authority to reject the result, access to sources and time to check them. Where no competent reviewer exists, narrow the use or do not use the output for a consequential task.

Never tell a patient, pupil, employee or family member that “the computer decided” when a person or organisation chose the system and acts on its output. Name who remains accountable and how the affected person can question or appeal.

Evidence practice: before using or sharing a digital output, write who or what produced it, the original source, the date, the intended use, the people affected, the personal data involved and the consequence if it is wrong. Check the claim against a named primary or authoritative source, inspect the surrounding context, look for independent confirmation and record uncertainty. For a religious claim, locate the complete Qur’an passage or authentic report and ask a qualified human scholar about its application. Do not cite an AI system as though it were the source of sacred knowledge.

5. Choose uses by consequence and reversibility

Low-consequence uses include generating private brainstorming prompts or reorganising text the user already understands. Medium-consequence uses include public summaries, translations or illustrations that could mislead an audience. High-consequence uses include diagnosis, legal rights, finance, safeguarding, religious rulings, identity, discipline or decisions affecting a person’s opportunities.

Increase verification and human authority as consequences rise. Ask whether an error can be detected, corrected and reversed; whether the person affected can contest it; and whether the source data were lawfully and fairly obtained. Convenience never turns an irreversible harm into a small risk.

Keep an output log for important uses: tool and version if known, date, prompt purpose, sources checked, edits made, reviewer and remaining uncertainty. This is documentation, not worship of the system. It enables correction when a model or source changes.

Digital-safety boundary: never upload passwords, private messages, confidential school or work files, intimate images, health records, children’s identifiers or another person’s data to a public AI service. Never use generated output as the sole basis for medical, legal, financial, safeguarding or religious decisions. Immediate threats, grooming, blackmail, self-harm material, sexual exploitation or credible violence require a trusted adult and the appropriate local professional or emergency route.

For adults teaching children: demonstrate with fictional prompts and non-animate sample media. Teach children never to enter a real name, face, voice, school, address, location, medical detail or private conversation into an AI tool. Keep adult-controlled accounts and privacy settings, supervise generated content, explain that convincing output can be wrong, and provide a simple route for reporting anything frightening, sexual, hateful, manipulative or secretive.

6. Write a personal technology covenant

Qur’an 99:7 promises that an atom’s weight of good will be seen. Technology can support learning, accessibility, communication, safety and useful service when the means and result are sound.

Qur’an 99:8 warns that an atom’s weight of evil will be seen. Scale can multiply a small dishonest act into harm for thousands; anonymity does not remove the deed.

Write: “I will not treat AI as divine or all-knowing. I will not ask it for unseen knowledge. I will verify important claims, protect other people’s data, disclose material AI assistance, refuse harmful imitation and keep a human decision maker responsible.” Add the trusted people and official sources you will consult.

Review the covenant monthly. Name one tool to remove, one notification to silence, one private dataset to protect and one beneficial task to improve. Return to the course whenever the technology or the intended use changes.

Evidence practice: before using or sharing a digital output, write who or what produced it, the original source, the date, the intended use, the people affected, the personal data involved and the consequence if it is wrong. Check the claim against a named primary or authoritative source, inspect the surrounding context, look for independent confirmation and record uncertainty. For a religious claim, locate the complete Qur’an passage or authentic report and ask a qualified human scholar about its application. Do not cite an AI system as though it were the source of sacred knowledge.