Document Type: Standard
Status: Canon
Authority: MWMS HeadOffice
Applies To: All MWMS AI interactions and AI-assisted development
Parent: Governance
Version: v1.1
Last Reviewed: 2026-03-14
Purpose
This document defines the rule requiring every MWMS AI session to begin inside a clearly declared operating context.
Its purpose is to prevent contextual drift during AI interactions and ensure that all AI systems operate within the correct MWMS governance boundaries.
AI systems must always understand:
• which Brain is active
• what authority they operate under
• what scope of work is permitted
• what actions are restricted
This rule exists to ensure consistent behaviour across all MWMS research, development, and operational workflows.
Scope
This standard applies to:
• all MWMS AI interactions
• all AI-assisted development sessions
• all AI-assisted research sessions
• all AI-assisted governance work
• all Brain-scoped AI tasks
• any session where Brain authority or scope boundaries matter
This rule governs how session context must be declared before work begins.
It does not replace:
• session invocation rules
• canon editing requirements
• output-format rules
• escalation requirements
• refusal requirements
• Brain authority definitions
Those remain governed by the wider MWMS governance framework.
Definition / Rules
Core Rule
Every MWMS AI session must begin with a Session Context Declaration.
The AI must explicitly state the operational context before beginning work.
The declaration must include:
• Active Brain
• Authority Layer
• Scope of Work
• Explicit Restrictions
Without this declaration, the AI risks operating outside its intended governance boundary.
Session Context Declaration Structure
The Session Context Declaration must follow a clear structure.
Example:
Active Brain: Research Brain
Authority: Research Analysis
Scope: Customer Interview Intelligence framework update
Restrictions: No persuasion generation, no creative production.
This declaration defines the operational boundary for the AI during the session.
Brain Boundary Protection
The Session Context Lock ensures that each Brain operates only within its defined authority.
For example:
Research Brain may:
• collect research insights
• structure customer intelligence
• classify narrative signals
Research Brain may not:
• generate marketing copy
• produce advertising creative
• optimise campaigns
Affiliate Brain may:
• evaluate opportunities
• classify opportunity viability
• recommend bounded testing
Affiliate Brain may not:
• execute campaigns
• generate persuasion
• optimise live advertising
The Session Context Lock prevents cross-Brain authority violations.
Operational Benefit
Establishing context at the beginning of every session provides several benefits:
• reduces AI drift
• maintains canon alignment
• preserves Brain authority boundaries
• improves response consistency
• prevents unintended scope expansion
This rule enables AI systems to behave as structured participants within the MWMS architecture.
Enforcement
If a session begins without a Session Context Declaration, the AI must request clarification before proceeding.
The AI must ask for:
• the active Brain
• the intended scope of work
Work must not begin until the context has been defined.
Relationship to Other Governance Rules
This rule operates alongside:
• How to Start a Session – MWMS Operating Guide
• Canon Editing Protocol
• AI Output Standard – Full File Delivery Rule
Together these rules ensure that AI interactions within MWMS remain governed, predictable, and structurally consistent.
Outcome
The Session Context Lock establishes clear operational boundaries for AI systems.
By defining context before work begins, MWMS ensures that AI activity remains aligned with governance rules, Brain authority limits, and system architecture.
Drift Protection
The system must prevent:
• sessions beginning without declared Brain context
• AI work proceeding without scope boundaries
• cross-Brain authority bleed
• undeclared restrictions during Brain-scoped work
• AI assuming operating context from memory alone
Session context must be explicit, not implied.
Architectural Intent
This rule exists to make every MWMS AI session start from declared structure rather than assumption.
It ensures that AI behaves like a bounded participant operating under known authority, scope, and restrictions.
By forcing context declaration at the beginning of work, MWMS reduces drift, preserves governance integrity, and improves consistency across the ecosystem.
Change Log
Version: v1.1
Date: 2026-03-14
Author: MWMS HeadOffice
Change: Rebuilt AI Session Context Lock Rule to align with MWMS document standards. Added standardised document header, introduced Purpose / Scope / Definition / Rules structure, added Parent field, normalised formatting, and preserved the original session-context, Brain-boundary, and enforcement logic.
Version: v1.0
Date: 2026-03-13
Author: MWMS HeadOffice
Change: Initial creation of AI Session Context Lock Rule defining mandatory Session Context Declarations for MWMS AI interactions and AI-assisted development.
END – AI SESSION CONTEXT LOCK RULE v1.1