M37
EN/DALet’s talk
Menu

AI Operating Model

Intelligence at every level.

M37’s three-layer operating model separates personal AI, functional AI and autonomous agentic AI—then connects all three through shared foundations and portfolio governance.

Design your operating modelExplore the transformation strategy
M37 / Working modelsStudy 05

One model. Three layers.

PersonalFunctionalAgentic
Support people in their everyday work.

What is an AI Operating Model?

An AI Operating Model defines how people, processes, technology, data and governance work together to select, build, operate and improve AI capabilities across an organization.

The three-layer model

One operating model. Three layers. Continuous value.

Start with human-led assistance, embed intelligence in functions, and introduce autonomous execution only where readiness and oversight support it.

01—03A three-layer model for scaling AI across the organization

Select a layer to focus the operating flow.

Help employees work faster and smarter with AI.

Examples
Meeting summariesContent creationResearch support
Key benefits
  • Higher productivity
  • Faster access to knowledge
  • Builds AI confidence
What success looks like

Employees use AI naturally in daily work.

Human-ledAI-assistedEmployee-owned

Embed AI into business functions and team workflows.

Examples
Procurement intelligenceQuality intelligenceCustomer-service intelligence
Key benefits
  • Better decisions
  • Faster processes
  • Greater consistency
What success looks like

AI supports departmental workflows and decisions.

Department-ledProcess-awareBusiness-focused

Deploy AI agents to execute and optimize workflows.

Examples
RFQ processing agentsCompliance monitoring agentsWorkflow orchestration
Key benefits
  • Faster execution
  • Scalability at scale
  • Improved resilience
What success looks like

AI agents execute workflows with human oversight.

Multi-agentWorkflow-drivenHuman oversight

Foundation · the operating-model enablers

F1Data quality & accessibility

Reliable, trusted data is the fuel for AI.

F2Integrated systems

Connected systems enable AI to act across the business.

F3Security & privacy

Protect data, systems and intellectual property.

F4Governance & ethics

Clear rules, risks, accountability and ethical AI use.

F5AI literacy & skills

Build AI fluency across all levels of the organization.

F6Change management

Drive adoption, engagement and continuous change.

Governance · AI & Automation Portfolio Board

01Prioritize initiatives02Allocate resources03Track value & ROI04Manage risk & compliance05Evaluate emerging technologies06Reassess & adapt

One operating model. Three layers. Continuous value.From individual productivity to autonomous enterprise operations.

Using the model

Make scaling decisions explicit.

Place each use case

Clarify whether value belongs at the personal, functional or autonomous layer.

Design the enablers

Identify the data, integration, security, skills and change capabilities each layer needs.

Govern the portfolio

Prioritize investments, track value, manage risk and adapt as technology changes.

M37

Design the way AI will actually operate.

We can map use cases to the right layer and define the enablers, controls and governance needed to scale them.

Talk to M37