Slide 1 of 12
Use arrow keys to navigate · Esc to close
SAF turns fragmented data into trusted decisions by detecting hidden relationships, behaviour patterns and trust risks before they spread across your organisation. Any domain. Any data. One engine.
Every entity passes through the same pipeline regardless of domain. Banking customer or hospital patient — the engine works identically. Powered by CO1–CO6.
That single question changes everything. Most data platforms were built to store and retrieve. SAF was built to understand, validate, and decide. That is a fundamentally different purpose — and it is why SAF is not limited to any single domain, industry, or data type.
SAF separates problem detection, severity classification, and governance response into three independent layers — making it truly domain-agnostic.
Gates detect specific semantic violations — a shared identity, a missing field, a deceased record still active in the system.
Each gate carries a weight tier that determines how severely the trust score is affected — from a gentle flag to a decisive block.
Override rules govern exceptional situations. P1 activates when a MANDATORY gate is detected — trust score bypassed, entity immediately BLOCKED.
SAF does not care whether the entity is a customer, patient, asset or vendor. The pipeline continues to operate. Only the mappings change.
Identify patient identity fragmentation, shared medical aid numbers, deceased active patients, and missing identity fields before they become clinical or financial risk.
Detect identity fraud, account collisions, dormant customer risk, and KYC gaps across your customer master — before onboarding or transaction approval.
SAF is domain-configurable. If your data has identity, relationships, and behaviour — SAF can govern it. No new engine required.
Healthcare and Banking are our first demonstration domains because both depend on trusted identity, relationship integrity, and accurate decision-making.
During development we discovered something important: the SAF engine itself does not change. Only the data mappings change.
Whether analysing patient records, customer identities, vendor networks, asset structures, or any other enterprise data, SAF follows the same process:
The domains shown here are demonstrations, not limitations. SAF is designed to operate wherever identity, relationships, behaviour, and trust matter.
SAF was not created by asking what was wrong with data.
It was created by asking what was wrong with how humans interpret data.
The result is a framework where meaning forms before decisions. Ten years of thinking — from Maslow to the Language Pyramid, from the Leadership Spiral to Semantic Governance. Read the origin story.
▶ Read the Origin of SAFNine signal detectors feed three coherence layers — aggregation, collision, and behaviour. Trust Decay adjusts for history. Priority rules check for mandatory blocks. CO4 issues the verdict. The Leadership Spiral is the philosophy that decided which questions were worth asking in the first place.
Organisations today hold millions of records but cannot confidently trust any of them. They do not know what is real, what is duplicated, what has drifted, or what has been silently corrupted over time. More storage and faster processing does not solve this. Understanding meaning does.
As computing evolves toward AI, graph intelligence and quantum systems, the challenge is no longer computation alone. The challenge becomes semantic coherence.
Most systems give you a snapshot. SAF gives you a story. Every entity has a complete trust history — every run, every decision, every change, every recovery. An auditor can replay the entire journey.
Interactive slide guides covering every gate, weight, signal, and decision layer. Read at your own pace — click to open, page through each deck in your browser. No video required.
Download a template, fill it with your own data, upload it and see SAF analyse it in real time. No account required.
Download the Excel template for your domain. Each template includes sample data with fraud patterns already injected.
Replace the sample records with your own. Add duplicates, test edge cases, or use real anonymised data.
Upload to the SAF portal. CO1 through CO4 runs automatically — trust scores, gates, and decisions in seconds.
Navigate the Entity Explorer, Relationship Graph, and Trust Engine. Every flag explained. Every decision traceable.
This guide contains detailed SAF architecture and research material.
Enter your Research Access Code to continue.
SAF is in private preview. We work directly with organisations ready to pilot semantic trust governance. Tell us what you're working with — we'll show you what SAF finds.