Back to logic overview

03 / Knowledge that reasons

Knowledge that
keeps its reasoning.

Connect sources, facts, rules and proofs in reusable knowledge. See how a changed rule affects existing answers and how one dependency supports a new question.

Start with this example
1A source changes2Related answers update3The proof stays attached

Your AI can build on what it has learned.

EXAMPLE 01 / A LIVING LOGIC BASE

One rule changes. Follow the consequences.

A connector lets AI access a data source. Its enablement policy gains one condition: a completed security review. Which connectors can still be enabled?

Choose a policy version, then supply the missing review fact for connector B.

Interactive illustration · Synthetic evidence
Policy documentPolicy v1 · source paragraph 3

Enable a connector if it is hosted in the EU and encryption is enabled.

Reviewed rule

A connector can be enabled when all of these conditions hold:

  • Hosted in the EU
  • Encryption is enabled
v1
Derived answers

A and B satisfy policy v1. C is outside the region condition.

2supported answers
ConnectorRegionEncryptedReviewResult
AConnector AEUYesCompleteConditions satisfied
BConnector BEUYesNot yet knownConditions satisfied
CConnector CUSYesCompleteOutside the EU condition
Inspect the formal rule+
enabled(C) ← region(C, EU) ∧ encrypted(C)

EXAMPLE 02 / ASK THE KNOWLEDGE A NEW QUESTION

One dependency. More than one answer.

A knowledge base links orders to components and components to a supplier. A new delay turns those existing relationships into a useful impact question.

Which orders depend on the delayed part?

S1Supplier SDelivery delayed
Sensor S
Order 41Order 42
Housing HOrder 43
Order 41Review delivery impact
Order 42Review delivery impact
Order 43No impact from this delay
Read the rule+
review(order) ← uses(order, part) ∧ supplied_by(part, supplier) ∧ delayed(supplier)

The records become more useful when their relationships can be traversed, combined and checked for a new purpose.

Store what made it true.

A knowledge base can retain why a conclusion holds, not just the words of the answer.

Sources

The original material and the exact supporting passage.

Facts

The value, subject, time and context extracted from that material.

Rules

The conditions, exceptions and reviewed version of the logic.

Proofs

The dependencies that connect a conclusion back to its premises.

A foundation that compounds.

Evidence → Reasoning → Reusable knowledge

01

Reuse across questions

Bring the same evidence and logic to a new answer, with its provenance intact.

02

Update with dependencies

When a source or rule changes, identify the affected conclusions and revisit their support.

03

Learn through review

Turn discoveries into reviewed knowledge, keeping their scope, examples and revision history attached.

CITPROOF

Build reasoning into your AI.

Bring us a workflow where answers, actions or discoveries need a stronger basis. Explore how evidence chains and logic verification can fit your product.

Talk to the team