The Operating Model Is Changing

For most of the history of the modern enterprise, people did the cognitive work.

1. People researched.
2. →People analyzed.
3. →People wrote.
4. →People coordinated.
5. →People made decisions.
6. →People supervised other people doing those things.

Software supported the work. It stored information, moved transactions and automated defined processes.

AI changes the role technology can play.

AI can interpret information, generate work, use tools, coordinate actions and execute sequences of tasks across systems.

That creates a new source of execution capacity inside the enterprise.

The operating model has to change with it.

When machines become participants in enterprise execution, the enterprise has to be designed for humans and machines working together.

That is the shift toward the AI Native Operating Model.

The strongest evidence is found in the work itself.

In a 15 September 2025 article, Microsoft described GitHub Spec Kit as an AI-based software development workflow in which teams outline project requirements, motivations and technical aspects before handing implementation to AI agents. Sentient Review interprets this workflow as a way for specifications to coordinate human direction and machine implementation, rather than evidence of a company-wide operating model.

In From Hierarchy to Intelligence, published on 31 March 2026, Jack Dorsey and Roelof Botha describe Block's proposed organizational model in which a company world model supplies context, directly responsible individuals own customer outcomes, and player-coaches combine building with developing people rather than primarily routing information. They state that Block is in the early stages of this transition and that parts may break before they work. Sentient Review reads this as an organizational design proposal, not evidence of a completed or successful transformation.

In a 12 November 2024 company release, Moderna identified Tracey Franklin as its Chief People and Digital Technology Officer, placing responsibility for people and digital technology under one executive title. Sentient Review reads the logic as direct. Workforce planning and technology planning become harder to separate when machines can perform parts of the work previously assigned to people.

A Walmart careers listing for a Senior Director, Software Engineering – Agent Builder, accessed on 2 November 2026, describes responsibility for the enterprise agent portfolio, a governance framework for autonomous agent deployment, the function's operating model and its organizational structure. Sentient Review reads this mandate as a sign that agent deployment is becoming an enterprise operating concern rather than a narrow software implementation. It is not evidence of an enterprise-wide outcome.

These companies are taking different paths.

That matters.

There is no settled AI Native organization chart.

What is becoming clearer is the problem they are responding to.

AI can now participate in execution.

The traditional workforce model was built around people.

A company estimated demand, translated that demand into work and determined how many people and systems were required to perform it.

AI introduces another form of capacity.

1. A customer service workflow can combine people with AI agents.
2. →An engineering workflow can combine developers with coding agents.
3. →A finance process can combine human judgment with machine analysis and execution.
4. →A sales process can combine relationship ownership with machine research, preparation and system updates.

The planning question changes.

Instead of asking:

How many people do we need to perform this work?

the enterprise can ask:

What combination of human judgment and machine execution produces the best outcome?

That question reaches far beyond workforce productivity.

It changes how work should be designed in the first place.

An enterprise may still have Engineering, Finance, Sales, Marketing and Customer Service.

The boxes can look familiar while the work inside them changes.

1. A developer may spend less time producing code and more time defining architecture, constraints and acceptance.
2. →A service employee may handle fewer routine interactions and more exceptions.
3. →A manager may spend less time collecting status and more time governing outcomes, capacity and exceptions.
4. →A finance professional may move from producing analysis toward evaluating assumptions and making decisions from machine generated work.

The title may stay the same.

The job underneath it may not.

This is why AI Native transformation begins with work.

Once work changes, other parts of the operating model follow.

1. Authority changes.
2. →Management changes.
3. →Skills change.
4. →Technology changes.
5. →Infrastructure demand changes.
6. →Economics change.

The evidence does not point toward one organizational template.

That is useful.

1. Block is testing changes to hierarchy and management.
2. →Microsoft is changing the relationship between specification and execution.
3. →Moderna has connected people and technology leadership.
4. →Enterprises across customer operations are assigning more routine execution to AI.

Each is testing a different part of the same question.

What should the enterprise look like when intelligence and execution are no longer supplied by people alone?

The answer will vary by industry, workflow, risk and economics.

But the design problem is now real.

The opportunity is larger than cost reduction.

AI creates the possibility of increasing the execution capacity available to an organization without requiring human effort to grow at the same rate.

1. Expertise can reach more work.
2. →Routine coordination can consume less human attention.
3. →Knowledge can move faster.
4. →Decision cycles can shrink.
5. →Organizations can absorb more demand.
6. →People can spend more of their time on judgment, relationships, ambiguity, creativity and ownership.
7. →Workflows can be rebuilt around outcomes rather than inherited handoffs.

None of these gains is automatic. Each depends on whether the work, the authority involved, the risk of the outcome and the economics of execution have been redesigned to support it.

That is the reason to pursue an AI Native Operating Model.

The goal is a more capable enterprise.

The change reaches across the organization.

1. Work needs a new allocation between humans and machines.
2. →Agents need defined authority.
3. →Human accountability needs clear ownership.
4. →Managers need to govern human and machine capacity.
5. →Enterprise context needs to be available to AI systems under controlled access.
6. →Agent activity needs identity, permissions and evidence.
7. →Infrastructure needs to support the execution demand created by the new work.
8. →Measurement needs to follow work through to accepted outcomes.
9. →Released human capacity needs a destination.
10. →Economics need to include both labor and machine execution.

These are connected design decisions.

Together, they form the AI Native Operating Model.

The first era of enterprise AI focused on a simple question:

Where can we use AI?

The next question is larger:

How should the enterprise be designed when both humans and machines can execute meaningful work?

The rest of this publication develops an answer.

It starts with the workflow.

1. Then the division of work.
2. →Authority.
3. →Management.
4. →Organization.
5. →Technology.
6. →Agents.
7. →Infrastructure.
8. →Economics.
9. →And the mechanisms required to move an enterprise from AI adoption toward an AI Native operating model.

The destination is a company with more capacity to execute, learn and adapt because human judgment and machine intelligence have been designed to work as one operating system.

That is the enterprise we now have the opportunity to build.