AI is not a feature. It is part of the workforce.

For years, software has helped property teams record work. It has held asset registers, service histories, documents, budgets, obligations and work orders.

But the work of connecting those records—understanding what has changed, what matters, who should act and what evidence supports the decision—has remained largely human. That is beginning to change.

AI can now read a report, identify a condition issue, connect it to the right asset, surface related obligations and prepare a maintenance proposal. But this only creates value when AI is part of the operating model: a governed member of the workforce, working from the same property reality, permissions, evidence and controls as the people responsible for the outcome. That is the principle behind Residention.

A workforce needs context

No experienced property professional makes a material decision from one isolated document. They consider the building, the asset, its condition history, the contract, the obligation, the budget, the people involved and the consequences of waiting.

AI should work the same way. An agent that can only see a prompt or a PDF may produce a useful summary, but it cannot reliably understand what that document means for a property operation.

Residention is designed to give people and governed AI agents a shared operating picture: the Property Twin. It connects the physical, operational, financial and regulatory reality of a property—its spaces, assets, documents, obligations, work, people, evidence and decisions.

An AI agent can then help monitor, interpret, prepare, coordinate and verify work within visible limits. It can show the evidence behind a recommendation, identify uncertainty and ask for clarification when material facts are incomplete. It can refuse where permissions, evidence or policy do not allow it to proceed.

Governance keeps people accountable

There is a major difference between an AI feature and an AI workforce. A feature answers a question; a workforce member has a role, a defined scope, access boundaries, a mission owner, a budget, a review path and a record of what happened.

In Residention, the work is designed to remain visible:

• The goal and the records and sources that may be used.
• The evidence supporting the recommendation.
• What is known, assumed or still missing.
• Who has authority to approve the next step.
• What the agent may do and what it must never do.
• How the work can be stopped, reviewed or recovered.

A recommendation, an approved decision, an executed action and verified completion are different states. They should never be silently collapsed into one.

AI can increase the capability of a property team. It should not obscure accountability.

People set policy, establish priorities and decide the appropriate balance between risk, cost, service, compliance and long-term value. They approve material decisions.

AI helps make those decisions better prepared. It can bring context together, reduce repetitive analysis, flag exceptions and prepare a source-backed proposal for the person with authority to decide. The objective is to give property professionals more time for work that requires judgement.

From individual tasks to an operating workforce

A property operation is full of repeatable work. These are not isolated tasks, but connected missions:

• An inspection report arrives.
• A duty approaches its due date.
• A supplier response is overdue.
• A meter reading moves outside its expected range.
• A planned intervention now affects the budget.
• Completion evidence needs to be checked and returned to the property record.

A governed AI workforce can help follow those missions across the operation: monitoring for change, gathering context, preparing work, coordinating the next step and ensuring evidence returns. Each action remains bounded, each decision remains visible and each exception has somewhere to go.

Over time, the operation becomes more capable not because it has added more software, but because its people and intelligence are working from the same truth.

Built for control before autonomy

Autonomy is not a switch to turn on because the technology exists. It has to be earned, process by process, through proven evidence, defined permissions, review gates, stop controls and safe testing.

::: Current status Residention provides the governed foundation for evidence-based AI agents. Production agent execution is currently disabled while isolated safety validation is completed. :::

That is deliberate. The first responsibility of an intelligent property operation is not to act faster, but to act on the right basis, with the right authority, in a way that can be understood and trusted.

What this means

The future property team will not be replaced by AI; it will be extended by it. People will continue to define strategy, make material decisions, manage relationships and exercise judgement, while governed AI agents help monitor, prepare and coordinate work within policy. That is not AI as a feature—it is AI as part of the workforce.