Assembling route context
Loading…
H-01: Night sky of drifting source fragments, slowly pulled toward the depth below.
Context General Intelligence
Every AI you run.One shared mind.
Your agents, copilots and workflows work from everything your organisation knows. Aware of each other. Each given exactly the context the moment needs. Loriq calls it Context General Intelligence.
Watch it happen / two agents, one mind
One agent hits a conflict. Every agent inherits the ruling.
Watch CORE take a live question through assembly and surface a contradiction for a human ruling. Then watch a second agent ask, and receive that ruling already packed in its context.
Agent A / staff assistantAgent B / onboarding workflowAuthored example / not customer data
Agent A / staff assistantWhat changed in our leave policy this quarter, and who approved it?
Stage 01 / InferInspectable
Understand the question.
CORE resolves the request, the relevant organisational domain and the time being asked about.
Domain / leave policy · intent / change + approval- Question scope
- Request date
- Organisational domain
Bounded context packet Receipts attached
Model-ready answer
The carry-over limit changed on 12 May. Leave Policy v3 supersedes v2, and the Quality Lead approved the change.
This is Context General Intelligence
Context General Intelligence is the capacity for every AI system an organisation runs to work from one shared, current and governed understanding of what the organisation knows, aware of each other, with every answer traceable to its sources.
Foundation models learned the world. They never learned your organisation. CGI is the missing half: the shared, governed understanding of what your organisation knows, working underneath every AI you choose to run.
Read the CGI thesisThe product
One shared mind for every model and agent
CORE is the platform behind Context General Intelligence. It holds what your organisation knows, keeps every connected tool current, and hands each request exactly the context it needs, with the receipts attached.
- 01 / CoordinationLIVE
What one agent does, every agent knows
H-03: A decision node pulses; agent nodes light in sequence as the position propagates. Fallback: all nodes lit. - 02 / Context assembly
Each request gets exactly the context it needs
H-04: A packet gauge fills exactly to the context budget, nothing more. Fallback: filled gauge. - 03 / Governance
Organisational controls apply before handoff
H-05: A gate closes over the packet; permission stamps land. Fallback: stamped packet. - 04 / Auditability
Every context package traces to its sources
H-06: A trace line draws backwards, answer → source → version. Fallback: fully drawn trace. - 05 / Sovereignty
You control the deployment and data boundary
H-07: A boundary ring draws itself around the whole diagram. Fallback: closed ring.
Active pilot / aged care
CORE is in an active pilot in Australian retirement living
The pilot partner is an Australian retirement living community. The pilot tests CORE against real organisational context: changing policies, authority, contradictions and source provenance.
Loriq is inviting a small number of organisations with a consequential context problem to become design partners.
Discuss a design partner pilotLORIQ / PILOT RECORDevidence, not adjectives
ActiveRetirement living community
Australia · retirement living
Context under testChanging policies, authority
Evidence handlingContradictions, provenance
EvaluationIn progress
Partner intakeDesign partners, limited
Loriq is not claiming outcomes while evaluation continues.
Make your organisation intelligible to every AI you choose.
Loriq / the missing half of intelligence is context