Open framework for AI-native operations
Own your operating model. Rent your tools.
The AI-Native Operating Framework is an open, vendor-neutral, interoperability-first business operating framework and method for defining, documenting, applying, and improving the standards and procedures through which people and AI perform work.
AI-native means people and AI may both perform business work under the same standards and SOPs. AI is not required in every process, and human accountability remains explicit.
The foundation is established and public. The detailed specification and worked examples are still being developed.
The problem
AI can accelerate tasks while the operating system falls further behind.
The framework starts with business work: how it is defined, owned, performed, controlled, reviewed, and improved when people and AI may both participate.
Individual speed is outrunning organizational coherence.
AI can make one person faster while the surrounding team still lacks shared standards, authority, evidence, and a dependable handoff.
Important work is trapped in people and tools.
The business loses continuity when operating context, decisions, process intent, and exceptions live only in chats, memory, or vendor-specific workflows.
Changing tools should not require organizational relearning.
The operating model should stay intelligible when models, platforms, vendors, teams, or implementation details change.
Framework core
Six concerns every business operation should make explicit.
The concerns define what must be clear without prescribing a lifecycle or organizing the framework around document types.
Intent
Purpose, scope, expected outcomes, and governing requirements.
Responsibility
Ownership, roles, authority, and accountability.
Work
Inputs, activities, outputs, dependencies, and handoffs.
Control
Policies, decisions, approvals, risks, exceptions, escalation, and recovery.
Assurance
Evidence, verification, quality, and completion.
Learning
Review, feedback, change, and continual improvement.
AI-native
One operating standard for people and AI.
AI-native means people and AI may both perform business work under the same standards and SOPs. AI is not required in every process, and human accountability remains explicit.
Core boundary
Business framework, not technology specification.
The framework governs business work and keeps its meaning portable across teams and changing tools. Models, harnesses, protocols, adapters, schemas, and machine-specific representations remain outside framework core.
SOP essentials
Eight areas of business meaning. No mandatory template.
Every SOP must make these areas clear. Organizations may combine, rename, or arrange sections to fit the work.
01
Purpose, scope, and expected outcome.
02
Accountable owner, participants, responsibilities, and decision authority.
03
Trigger, prerequisites, inputs, and authoritative sources.
04
Activities, decision points, dependencies, handoffs, and outputs.
05
Applicable policies, controls, approvals, and risks.
06
Exceptions, escalation, recovery, and stop conditions.
07
Completion criteria, verification, and required evidence.
08
Review ownership, review cadence or trigger, and change history.
Keeping standards current
Understand → Document → Validate → Approve → Use → Improve
These activities maintain operating standards. They do not prescribe the lifecycle of the business process being described.
01
Understand
Observe the actual work and gather governing requirements, evidence, and practitioner knowledge.
02
Document
Describe the standard or SOP using the framework concerns and content requirements.
03
Validate
Walk through normal work, decisions, handoffs, exceptions, and failure scenarios with the responsible people.
04
Approve
Obtain approval from the accountable business owner and any required policy or control authorities.
05
Use
Make the approved standard available, communicate changes, and perform the work according to it.
06
Improve
Review outcomes, evidence, exceptions, incidents, and changed requirements; revise and reapprove when warranted.
Founding principles
Interoperability starts with shared organizational understanding.
These principles keep the operating model clear while teams, tools, vendors, and implementation details change.
Interoperability first
Teams share a clear organizational understanding and consistent ways of working across business boundaries.
Tool-independent operations
Business standards stay canonical while tools, models, and vendors remain replaceable implementation choices.
Process before platform
Define the work, owner, authority, evidence, and controls before choosing how technology participates.
One body of documentation
People and machines use the same clear business standards and SOPs rather than separate machine-specific rules.
Human accountability
AI participation never removes accountable human ownership of business outcomes.
Evidence and recovery
Work should be observable, reviewable, reversible where practical, and explicit about exceptions and recovery.
Progressive adoption
Organizations can apply the framework to existing processes without replacing every sound operating practice.
Open development
The method improves through real use, shared examples, and review.
Design choices
Open enough to travel. Clear enough to use.
The framework keeps business meaning stable without forcing one vendor, document layout, or implementation path.
Open and usable without Digital Meld
Centered on business work, not one software-delivery lifecycle
Independent of any model, harness, protocol, schema, or vendor
One operating standard for people and AI
Compatible with sound processes an organization already uses
Improved through real use, review, and shared learning
How to use it
Learn the method, practice it, review the work, or get implementation help.
Each path stands on its own, so organizations can start where the need is clearest.
How the pieces fit
Start with the method, learn it through real workflows, review completed work, or get implementation help.
Open product + method
AI-Native Operating Framework
Defines a shared method for making business work clear, accountable, and portable across teams and tools.
Education + workshops
AI Dev Days
Turns the framework into practical talks, labs, scenarios, and facilitated workshops.
Case studies + public work
Work
Shows completed work, implementation decisions, results, and the public artifacts behind them.
Commercial application
Digital Meld
Helps organizations assess, implement, train, migrate, govern, and operate real business work.
Brad Groux created and stewards the framework. Digital Meld helps organizations apply it to real operating work.
Illustrative domains
Ten areas where the framework will be demonstrated.
Detailed examples are still being developed. These domains show the intended breadth; they are not finished case studies or requirements for using the framework.
01
Accounts-payable invoice processing
02
Software-change delivery
03
Construction field-incident response
04
Employee onboarding and offboarding
05
M&A Day 1 transition
06
Customer complaint and service recovery
07
Regulatory-change implementation
08
Supply-chain disruption response
09
Sales proposal and contract approval
10
Patient referral and care transition
Learn the open method. Apply it where the work is real.
Practice through AI Dev Days, review the public work, or bring Digital Meld a workflow that needs accountable implementation.