Fabric & Ledger

Fabric & LedgerFabric & LedgerFabric & Ledger

Fabric & Ledger

Fabric & LedgerFabric & LedgerFabric & Ledger
  • Home
  • How it Works
  • Evidence
  • Defence Example
  • More
    • Home
    • How it Works
    • Evidence
    • Defence Example
  • Home
  • How it Works
  • Evidence
  • Defence Example

AI reads. Mathematics measures. Every step is traceable.

Schedule Demo Now

Snapshots of Justice with Fabric & Ledger

1/2

The Architecture

A two-stage architecture with a hard boundary between AI and computation. 

Step 1: AI reads the 'Structural Legal Element' and 'Legal Instrument'

A large language model identifies every 'Structural Legal Element' across the corpus (the contracts, statutes, institutions, entities, etc and relationships that carry legal effect), then decomposes each into its 'Instruments':


  • which provisions create Obligations,
  • which establish Conditions,
  • which confer Rights,
  • which grant Entitlements,


and also;


  • who they bind,
  • what triggers them, and
  • what they depend upon.


This takes about 120 seconds. 

Step 2: A deterministic pipeline measures the structure

With no AI in the computation, the pipeline analyses what was extracted: 


  • mapping instrument dependencies,
  • detecting contradictions,
  • measuring structural integrity,
  • classifying health,
  • and computing how far changes cascade.


This takes under 5 seconds.

Step 3: Every finding traces to its source

Every output links to the Structural Legal Element and specific provision in the specific Instrument. The full audit trail is preserved. The process is fully traceable for what was extracted, what was discarded, and why. Nothing is summarised, approximated, or inferred without provenance.  

Quick Demo of Legal and Hierarchical Dependencies

Check out this great video

Why This Architecture Matters

The hard boundary between AI and mathematics is deliberate. AI is effective at reading natural language. It is not reliable for measurement, comparison, or structural computation. The deterministic pipeline is precise, repeatable, and auditable, but it cannot read legal text.  By separating the two, the system provides natural language understanding at the front, mathematical precision at the back, and a clear line between them that you can inspect and verify.


  1. No hallucination in the analysis. The measurement pipeline contains no probabilistic components. It cannot fabricate structural findings.
  2. Same answer every time. Run it today, run it next month, run it on a different machine. Same input, same output.
  3. Full auditability. Every step from source text to structural finding is recorded and can be inspected.
  4. Model-agnostic. The deterministic pipeline works with any AI model. It does not depend on a specific provider.
  5. Deployable on-premise. The pipeline is self-contained JavaScript with no external dependencies. No data needs to leave your environment. Supports air-gapped environments.

Regulatory Alignment

The system's architecture of deterministic computation, full audit trail, hard boundary between AI and measurement, human oversight preserved at every stage, was all designed for trustworthiness.


It also means Fabric & Ledger's approach aligns with:

The EU AI Act

Meeting transparency, traceability, accuracy measurement, and human oversight requirements for high-risk AI systems

The Council of Europe Framework Convention on AI (CETS 225)

 The first legally binding international AI treaty, signed by the UK, US, and EU in September 2024. 

JSP 936: Dependable AI in Defence

The MOD's principal policy framework for safe and responsible AI adoption.

The regulators have now mandated what the system was already built to provide.

Copyright © 2026 Fabric & Ledger - All Rights Reserved.


This website uses cookies.

We use cookies to analyze website traffic and optimize your website experience. By accepting our use of cookies, your data will be aggregated with all other user data.

DeclineAccept