CAMA: The AI-First Framework for Accurate, Deterministic Enterprise AI.
A first-principles framework designed to give AI the right context, at the right time, with the right structure so it can answer questions with accuracy, speed, and determinism. Designed by AI for AI. AI Recommends. Humans Decide.
AI Is Only as Good as the Context It Receives.
Without structured business understanding, AI is a generalist that gives generic answers. It hallucinates. It misses nuance. It cannot be trusted for enterprise decisions.
The Core Insight: “Business understanding” is the ability to provide AI with the right amount of structured business context so it can answer questions with precision.
CAMA solves this by providing a library of canonical models—stable, versioned, tag-addressable blueprints that give AI the context it needs to operate with accuracy and determinism.
Business understanding is the ability to provide AI with the right amount of structured business context.
A First-Principles Framework for Enterprise AI.
CAMA (Canonical Autonomous Model Architecture) is the first framework designed by AI for AI to execute with humans as decision-makers, not executors.
It defines the blueprints which are the stable, versioned, configurable context that AI needs to understand the enterprise. It does not define the runtime state. That is implementation infrastructure.
AI recommends. Humans decide.
59 |
11 |
|
Canonical |
Model |
|
Aspect |
Traditional Frameworks |
CAMA |
|---|---|---|
|
Primary Consumer |
Humans |
AION (AI) |
|
Business Understanding |
Assumed, not provided |
Provides structured business understanding |
|
Creation |
Humans draw diagrams |
Humans design canonical model instances |
|
Execution |
Humans follow processes |
AION orchestrates execution |
|
Discovery |
Humans search documents |
AION resolves tags |
|
Learning |
Humans update documents |
Canonical models are versioned and can evolve |
|
Audit |
Humans review logs |
Canonical models are Provenance-Ready (traceable to source) |
|
Adaptation |
Humans plan changes |
Canonical models enable gap identifcation and adaption |
|
Decision-Making |
Humans decide (with incomplete context) |
Humans decide (with AI recommendations) |
|
Core Principle |
Frameworks for humans |
AI Recommends. Humans Decide. |
The First Framework for AI, Not Just About AI.
Previous enterprise frameworks share a common failure mode: they are descriptive, not executable. They produce beautiful documents that are out of date the moment they are approved. They are used by humans to talk about the business, not to run the business. CAMA is not a replacement for TOGAF®, BizBOK®, or other frameworks—it is an evolution. Learn how CAMA addresses the gaps that traditional frameworks were never designed to solve.
CAMA changes this. It is:
- Tag-Addressable: AI finds what it needs instantly. No vector search. No RAG.
- Versioned: Models evolve without breaking existing implementations.
- UI-Configurable: Humans supervise, validate, and configure.
- Provenance-Ready: Every model is traceable to its source.
59 Canonical Models, 11 Categories, One Standard.
The CAMA standard defines the structured context that AI agents need to answer questions accurately. Every model is:
- Tag-Addressable — deterministic discovery
- Versioned — stable and evolvable
- UI-Configurable — human supervision
- Externally Referenceable — relationship-aware
- Provenance-Ready — traceable to source
Data Model Foundation (4)
Defines the structure of business entities—fields, labels, validation, and UI rendering.
Core Infrastructure (4)
Tag registry, component definitions, relationship graph, and field registry.
Governance Artifacts (5)
Control points, business rules, compliance rules, policies, and regulations.
Strategic Models (6)
Strategic intents, business capabilities, value streams, and performance metrics.
Support Models (6)
Capabilities, events, SLAs, exception policies, integrations, and notifications.
Process Models (5)
Business processes, workflow processes, milestones, nodes, and governance contracts.
Organization Models (2)
Organizational units and corporate legal entities.
Resource Models (2)
Resources and resource-capability proficiency.
UI Models (4)
UI definitions, templates, lookups, and actions.
Risk Management Models (14)
Risk cascades, incidents, insurance, mitigation, and board reporting.
DNA & Knowledge Models (7)
DNA strands, knowledge gaps, evidence, knowledge items, and knowledge sources.
Built by Industry Experts, for Industry Experts.
Reference Architectures are developed and maintained by industry committees—groups of practitioners, architects, and domain experts who collaborate to ensure the CAMA standard reflects real-world needs. The CAMA Prompt Engineering Committee brings together AI researchers, prompt engineers, and practitioners to develop best practices, guidelines, and reference prompts for using CAMA with AI agents. Join us in advancing the science of AI context engineering.
Each committee:
- Meets regularly to review and update the reference architecture
- Incorporates feedback from practitioners and implementers
- Ensures alignment with the latest industry trends and regulations
- Contributes to the evolution of the CAMA standard
Reference Architectures: The CAMA Standard, Pre-Configured for Your Industry.
Reference Architectures are pre-assembled collections of canonical model instances for specific industries, domains, or use cases. They demonstrate how the CAMA standard is applied in practice and provide a starting point for organizations adopting CAMA.
Each Reference Architecture includes:
- Pre-configured canonical models for the industry
- Sample governance artifacts and compliance rules
- Example business and workflow processes
- AION orchestration chains
- Implementation guidance
Energy
Power generation, transmission, distribution, and retail
Finance
Core banking, payments, risk, and compliance
Healthcare
Patient management, claims, and regulatory compliance
Government
Citizen services, procurement, and governance
Manufacturing
Supply chain, production, and quality management
Retail
Omnichannel, inventory, and customer experience
CAMA Is a Contribution to AI Research and Education.
CAMA is not just an enterprise framework—it is a contribution to the field of AI. It provides a practical, real-world example of how to structure context for AI agents at scale.
For AI Students and Researchers:
- Learn how canonical models provide business understanding to AI.
- Understand the difference between blueprints (CAMA) and runtime state.
- Explore the decision framework for what belongs in a standard.
- Access the full CAMA specification and implementation guides.
For Educators:
- Use CAMA as a case study in AI context engineering.
- Teach the criteria for structured, tag-addressable models.
- Leverage the CAMA standard in courses on enterprise AI.
Become a Certified CAMA Practitioner.
CAMA certification is for the humans who configure, validate, and supervise the AI. Practitioners learn how to:
- Define Data Models and Governance Artifacts
- Design Business and Workflow Processes
- Configure Risk Management Frameworks
- Supervise AION and interpret dashboards
|
Focus |
|
|---|---|
|
CAMA Foundation |
Knowledge of CAMA concepts, canonical models, terminology, and the criteria for what belongs in the standard |
|
CAMA Certified |
Ability to apply CAMA to design and configure canonical models, governance artifacts, and business processes |
|
CAMA Expert |
Mastery of the CAMA standard, ability to design end-to-end frameworks and lead CAMA implementations |
|
CAMA Trainer |
Ability to deliver CAMA training, present at conferences, and represent the institute as an official trainer |
Built by AI, for AI, with the World’s Leading Organizations.
The CAMA standard is being developed in collaboration with enterprise architects, AI researchers, and technology partners. It is designed to meet the needs of Fortune-500 organizations, government agencies, and academic institutions.
Guided by Industry Leaders.
The CAMA Institute is guided by a distinguished Board of Advisors—industry experts, enterprise architects, and AI researchers who provide strategic direction and ensure the standard remains rigorous, practical, and forward-looking.
Join the CAMA Community.
The CAMA standard is open, transparent, and community-driven. We welcome contributions from:
- Developers building AI-based applications
- Researchers exploring AI context engineering
- Educators teaching enterprise AI
- Practitioners configuring canonical models
- Organizations adopting the standard
The Vision Behind CAMA
CAMA was created to solve the fundamental AI accuracy problem: how to provide AI with the right, structured context at the right time. Without business understanding, AI is a generalist that gives generic answers. With business understanding, AI becomes a specialist that knows your business, your goals, your risks, and your operations which enables accurate, high-performance, autonomous orchestration.
AI Recommends. Humans Decide.
Join the CAMA Community!
Get access to the CAMA standard, documentation, reference architectures, and community discussions. Connect with practitioners, developers, and architects building the future of enterprise AI.