TasnidChain combats knowledge decay and AI hallucinations by grading claims based on their transmission chains — the scholarly tradition of isnad applied to the digital age.
Information degrades as it passes through unreliable channels. Context is lost, nuance disappears, and errors compound. Without rigorous transmission tracking, knowledge loses its authenticity.
Large language models generate plausible-sounding but false information. They predict text probabilistically, not truth. Without a verification layer, AI-generated content pollutes the information ecosystem.
We grade knowledge claims based on their transmission chains — who said it, who verified it, and how reliable are the transmitters. The 1,400-year-old isnad methodology, digitized for the modern world.
TasnidChain transforms unverified claims into structured, graded knowledge with full provenance.
Unverified Knowledge
Verified Knowledge
Every claim is evaluated based on its chain of transmission, following the scholarly tradition of isnad.
Enter a knowledge claim with its source and domain. New submissions start as T1 (Unverified) and enter the verification pipeline.
Trace the transmission chain — who originated the claim, who transmitted it, and their reliability scores. Each link is recorded and weighted.
Claims receive a trust grade (T0-T5) based on chain strength, transmitter reliability, and cross-domain corroboration. Full computation is transparent.
Every claim receives a composite trust score computed from three weighted dimensions.
90-100%
Verified by scholarly consensus with strong transmission chains
80-89%
Supported by reliable transmitters and substantial evidence
65-79%
Supported by evidence with some gaps in transmission
45-64%
Limited evidence or transmission chain weaknesses
<45%
Significant disagreement or weak evidence
Flagged
Demonstrated to be false or fabricated
Explore claims across different domains, each with full transmission chains, trust grades, and transparent composite scores.
LLMs generate plausible text without knowledge of truth. TasnidChain gives AI systems a way to verify claims before presenting them to users.
Before an LLM answers a question about Islamic knowledge, it queries TasnidChain to verify the claim and get its trust grade.
AI responses include TasnidChain trust grades (T0-T5) so users know how much confidence to place in each claim.
Every verified claim includes its complete transmission chain — who said it, who transmitted it, and their reliability scores.
Integrate TasnidChain into any AI pipeline. LLM companies pay for verified Islamic knowledge API access — the primary revenue model. Every response includes full provenance and trust grades.
GET /api/claims — Query verified claims by domainGET /api/claims/:id — Full claim with chain detailsPOST /api/verify — Submit a claim for gradingGET /api/narrators — Transmitter reputation dataGET /api/graph — Knowledge graph traversal// LLM sends claim for verification
{
"claim": "Fasting in Ramadan is obligatory",
"domain": "QURAN",
"source": "Al-Baqarah 2:183"
}
// TasnidChain responds
{
"trustGrade": "T0",
"compositeScore": 0.96,
"transmitterScore": 0.97,
"chainScore": 0.95,
"corroborationScore": 0.94,
"verdict": "Consensus Verified",
"chains": [{ "links": 4 }]
}