> ## Documentation Index
> Fetch the complete documentation index at: https://docs.integratewise.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Human + AI Operating Model: Authority, Collaboration, Continuity

> In SpineWorkspace, AI does the heavy lifting and humans stay on top with intent, judgment, and approval. Same goal, one journey.

SpineWorkspace is built around a specific operating model. AI does the heavy, monotonous, tedious work. Humans stay on top with intent, judgment, and approval. Both operate on the same shared business context. Both move toward the same business goal. This is not AI replacing people. It is AI and people on one journey, with human authority preserved at every step.

## The Division of Labor

The split is not about capability. AI can draft, review, propose, and execute. Humans can too. The split is about where each adds the most value and where authority must live.

<CardGroup cols={2}>
  <Card title="Human Role" icon="user">
    Set intent and direction. Apply judgment on ambiguous cases. Approve or reject proposals. Escalate when context demands it. Own the outcome. Remain accountable.
  </Card>

  <Card title="AI Role" icon="robot">
    Assemble context from every connected system. Draft responses, prep docs, and recommendations. Review large volumes for patterns and anomalies. Execute governed capabilities under approval. Learn from receipts to improve the next proposal.
  </Card>
</CardGroup>

## The Shared Journey

Work in SpineWorkspace follows a continuity loop. The human and AI move through it together, each contributing what they do best.

<Steps>
  <Step title="Observe together">
    Spine Fabric assembles live context from every connected system. The human sees the projected view. Twin and domain agents see the same unified context. Nobody starts from zero.
  </Step>

  <Step title="Decide together">
    The AI proposes based on the full picture: draft a message, flag a risk, suggest a next step. The human reviews the proposal, asks for clarification if needed, and applies judgment the AI does not have.
  </Step>

  <Step title="Act with approval">
    Execution flows through governed capabilities. The AI prepares the action. The human approves. The write happens. A receipt is generated. Nothing hits a system of record without human authority.
  </Step>

  <Step title="Learn from receipt">
    The outcome becomes an event in the shared continuity layer. The business memory grows. The next cycle starts richer. Agents learn patterns. Humans learn what to trust.
  </Step>

  <Step title="Continue">
    The loop repeats. Context compounds. The team gets faster and more confident because the shared memory deepens with every iteration.
  </Step>
</Steps>

## Two Workbenches, One Context

Humans use SpineWorkspace as their operational surface. AI agents use the same Spine Fabric as their reasoning substrate. The workbench looks different for each role, but the context underneath is the same. This symmetry is what separates SpineWorkspace from tools that bolt AI onto a sidebar and hope the user feeds it enough context to be useful.

## Human Authority Is Non-Negotiable

In this model, AI never acts in place of human authority. It acts under it. An agent can propose, draft, review, and even execute within a pre-approved scope. But governance sits between operational work and canonical state. Write-back to source systems requires authorization. Exceptional cases require human escalation. The accountability chain always terminates with a person.

<Note>
  Preserving human authority does not mean preserving inefficiency. Most approvals in SpineWorkspace are one-click because the AI assembled the context and the reasoning transparently. The human is informed, not burdened.
</Note>

## Why This Model Works

* It leverages AI for scale and speed without discarding human judgment.
* It keeps governance simple because authority is always clear.
* It builds trust over time because humans see what the AI saw and why it proposed what it did.
* It compounds organizational knowledge because every approved action deepens the shared memory.

IntegrateWise designed SpineWorkspace around this operating model because the future of work is not humans versus AI. It is humans and AI, on the same context, toward the same goal, with human authority intact.

## Related

* [Shared Context Principle](/product/shared-context-principle)
* [Canonical vs Operational](/product/canonical-vs-operational)
* [SpineWorkspace](/product/spine-workspace)
* [Twin](/product/twin)
* [Governed Actions](/how-it-works/governed-actions)