Admissible Evidence Spans

This concept explains how AI systems prove they read the right documents before answering your questions. In fields like finance, law, or medicine, an answer is only useful if you can trace it back to a specific, verified source document that existed at a known time. If you cannot point to the exact text that supports a claim, the answer is not trustworthy in a professional setting. Understanding this mechanism lets you build systems where every statement made by an AI can be checked against the original records, turning vague responses into verifiable facts.

Imagine you are a junior analyst at a financial firm. A senior partner asks you a question: 'What were the risks mentioned in the last quarterly filing of Company X?' You open the firm's document system and search. You pull up several documents, each with a few paragraphs that might answer. But you cannot just paste the first thing you find into your reply. You must point to the exact paragraph — the precise span of text — that supports each claim you make. That exact paragraph is an 'admissible evidence span'. It is admissible because it comes from a verified source (the filing), and it is a span because it is a specific, bounded section of text, not the whole document or a vague reference.

Now think about how a modern AI assistant works when you ask a question. The system does not just 'remember' the answer. It finds relevant documents, identifies passages within them, and generates an answer based on those passages. If the answer is grounded in admissible evidence spans — meaning each claim is backed by a specific, traceable passage from a trusted source — then you can click a number in the answer and see the exact sentence that supports it. If the passage is not admissible — say it comes from an unverified blog or a stale version of the filing — the whole answer becomes suspect. In regulated settings, the difference between 'admissible' and 'inadmissible' can be the difference between a compliant answer and a lawsuit.

Think of a courtroom trial. A lawyer cannot just say 'the defendant was at the scene' — she must present evidence: a witness who saw it, a CCTV clip, a signed document. That evidence must be admissible: it must be authentic (not fabricated), relevant (it actually speaks to the case), and properly obtained (not hearsay or illegally gathered). And it must be specific — you cannot cite 'the whole investigation file'; you point to exhibit A, page 3, lines 12–14. An AI system that answers questions is like a lawyer building a case. The 'admissible evidence spans' are the specific passages it may cite. The system's search step is like the lawyer's investigator pulling relevant documents. But the lawyer must check admissibility: is this passage from a trusted source? Is it about the question? Is it specific enough to pinpoint? Only if it passes these checks can it be used in the final argument — the AI's answer.

An admissible evidence span is a specific, continuous section of text from a verified document that an AI system is allowed to use and cite as proof for a claim in its answer. It must come from a trusted source, be relevant to the question asked, and be precise enough to identify exactly where the information came from.

This concept sits next to Retrieval-Augmented Generation, which is the practice of giving AI models access to external documents so they can answer questions using that text rather than just their training memory. It is also closely tied to Chunking, which is the process of splitting large documents into smaller, manageable pieces so they can be stored and searched effectively.