TafsirPrism

Sources & Method

TafsirPrism's only reason to exist is trust: we openly share where every text we show comes from, under what license, and how the AI works with these texts.

Sources

The Arabic Quran text and surah/juz/page data come from the Tanzil Project (tanzil.net) — openly licensed, used unmodified, with attribution.

Translations: Turkish by Elmalılı Hamdi Yazır (public domain), English by M. M. Pickthall (public domain), German by Bubenheim & Elyas (via Tanzil, under non-commercial terms) — all three via Tanzil's digital edition.

Tafsirs

Tafsir texts come from the OpenITI Corpus (openiti.org) — an academic open-access project digitizing classical Islamic texts. License: CC BY-NC-SA 4.0, non-commercial use only. That's why TafsirPrism is currently donation-supported/free.

Abu al-Su'ud and Elmalılı will be added later. Some tafsir sources (e.g. altafsir.com's own digital collection) were deliberately not used because they would have violated that site's terms of use.

How does the AI work?

  1. Retrieve: Passages from each tafsir are located for the relevant verse.
  2. Summarize: For each tafsir separately — a view, its basis, and a verbatim quote are produced.
  3. Verify: Is the generated quote really a verbatim match in the source text — checked in code. If not, it's rejected; the model isn't trusted blindly.
  4. Compare: Built only from verified summaries — common ground, differences, historical context.
  5. Modern connection: Written cautiously, without claiming a final conclusion.
  6. Safety filter: If absolute language or sensitive topics are detected, the content goes to human review.

Why free?

Our tafsir sources' license (CC BY-NC-SA) excludes commercial use. That's not a deliberate choice — it's the best option currently available. Our own digitization work or direct permission could change this later.