~1,400 years ago, scholars built a rigorous system to verify who you can trust. I rebuilt it as a trust layer for AI agents.

I wrote this and just put it on arXiv, sharing for the discussion. When statements spread through long chains of people — some reliable, some not — you can't trust a claim just because it sounds right. Islamic scholars faced this centuries ago and built one of history's most rigorous systems for verifying transmitted knowledge: every claim carries its full chain of transmitters (isnād), every transmitter is graded on integrity and precision (rijāl), the chain is only as strong as its weakest link, independent chains raise confidence, and even a flawless chain doesn't excuse a flawed message. Now look at AI in 2026. An answer passes through a scraper, an extractor, several models, a synthesizer. Some links are reliable, some aren't — and when they fail, they fail silently. A confident, fluent answer that's quietly wrong. Everyone is racing to verify the agent: its identity, its permissions, its access. Almost no one is verifying the claim: whether what it said is true and independently corroborated. So I took that centuries-old methodology and rebuilt it as a trust layer for multi-agent AI. I call it ISNAD. Everyone verifies the agent; ISNAD verifies the claim. The rigor belongs to twelve centuries of scholars — the transfer to AI is mine. I also wrote the failures into the paper: some mechanisms are validated, others aren't yet, and I said so in detail. A trust framework that hides its weaknesses is a contradiction in terms. Paper: https://arxiv.org/abs/2607.24117 Code: https://github.com/alizahidraja/isnad Agree or disagree, I'd love to hear it.