What does Orbital Witness use AI for in property due diligence?

Orbital Copilot performs commercial legal-document analysis and information gathering for property due diligence and source-linked reports, using Microsoft Azure OpenAI. The source establishes domain evaluation and customer use, but not autonomous runtime tool control or completion of a property transaction.

Who checks Orbital Witness's AI output?

The source describes legal engineers comparing prompts with expected outputs and supplying source references in test environments. It does not name law-firm execution details. As a proposed design, the agent drafts and a qualified lawyer checks interpretation and advice, with no delegated sign-off.

How strong is the evidence for Orbital Witness's product?

Evidence is early-stage. The published account describes early law-firm deployments followed by a commercial product, so actual use is established. Named law-firm execution details, external tool boundaries and independently measured legal accuracy are absent.

What would an enterprise need to build something similar?

This is Nuvepro's proposed design, not Orbital's. It would reload an approved anonymized document bundle with a gold-standard issue set, include cases such as a missing annex or conflicting title record, and test issue recall, unsupported conclusions and escalation to qualified counsel.

Limited published evidence · Legal technology

AI property due diligence: Orbital Witness example

Orbital Witness reports Orbital Copilot, running on Microsoft Azure OpenAI, which analyzes commercial legal documents and gathers information for property due diligence, producing source-linked reports. Legal engineers develop prompts and assess expected outputs in test environments. Autonomous tool control is not established.

Property due diligence and source-linked reports

Orbital Witness

Published account: Early law-firm deployments followed by commercial product; actual use is established.

A published third-party example, not a Nuvepro customer case or evidence of a partnership.

The workflow at a glance

Orbital Witness reported work

Illustrative summary of the published account

Reported work components. Their arrangement does not imply a fixed execution order or simultaneous work.

Legal product
Orbital Copilot
Commercial-document analysis and information gathering support property diligence.
Domain development
Legal-engineer evaluation
Prompts and expected outputs are assessed in test environments.
Work output
Source-linked diligence reports
The account describes targeted reports and encourages customers to check difficult parts.

Published account: Early law-firm deployments followed by commercial product; actual use is established. Named law-firm execution details, external tool boundaries and independently measured legal accuracy are absent. See the source account and limitations below.

What the source reports

Orbital Copilot performs commercial legal-document analysis and information gathering. Legal engineers develop prompts and assess expected outputs in test environments. The source establishes domain evaluation and customer use, but not autonomous runtime tool control or completion of a property transaction.

Reported technology: Microsoft Azure OpenAI. Scope and outcomes are source-reported.

What the enterprise environment would need

The following is a proposed project design for an enterprise doing similar work. Published details are marked; the remaining systems, data and scenarios would be agreed for your project.

Proposed practice environment

Prepare a property diligence report with source evidence

Illustrative project design to scope with your enterprise.

1 · Set up the environment
Property document bundle
Fictional titles, annexes, clause versions and source locations retain unresolved issues.
2 · Preparation and agent contributions
Draft diligence findings
Gather information and prepare source-linked property risks or gaps.
3 · Independent checks
Independent lawyer checks
Reference issues test recall, unsupported conclusions and missing annex handling.
4 · People decide acceptance
Qualified lawyer accepts draft
Review interpretation and uncertainty while retaining legal advice responsibility.

Only within the accepted practice scope

Practice output

A reviewable evidence package

A diligence report, issue list and source-provenance register.

If a check fails → withhold acceptance.

A missing annex or conflicting title record is concealed by a fluent conclusion. Withhold acceptance, repair, independently recheck and return to owner review.

Failure or owner rejection returns to review

Hold
→ 2 · Revise work
→ 3 · Rerun independent checks
→ 4 · Owner review

Operating boundary. Controlled legal corpus only; no conveyancing action, client advice or delegated legal sign-off.

Proposed, not an executed Nuvepro project or a description of the named company’s implementation. Owners agree the data, rules, responsibilities and acceptance criteria for their enterprise.

Systems and tools: Inferred test scope: Document corpus and legal analysis tools; no replicated conveyancing suite required initially.
Data and state: Inferred test scope: Document version, clause provenance and unresolved issue state.
People and handoffs: Inferred test scope: Agent drafts; qualified lawyer checks interpretation and advice.
Rules and authority: Inferred test scope: No unsupported legal conclusion or delegated sign-off.
Normal and exception cases: Proposed test cases: Missing annex, conflicting title record, handwritten clause, outdated document.
Reset and replay: Not reported. Proposed: Reload an approved anonymized document bundle and gold-standard issue set.

What needs checking

Published: legal engineers compare prompts with expected outputs and supply source references.
Proposed: test issue recall, unsupported conclusions and escalation to qualified counsel.
Proposed: qualified lawyers compare represented documents and assumptions with real diligence practice.

Review environment fidelity, human capability and agent performance separately. The enterprise’s accountable owners decide acceptance of the shared workflow and any operating use.

Questions about this example

What does Orbital Witness use AI for in property due diligence?: Orbital Copilot performs commercial legal-document analysis and information gathering for property due diligence and source-linked reports, using Microsoft Azure OpenAI. The source establishes domain evaluation and customer use, but not autonomous runtime tool control or completion of a property transaction.
Who checks Orbital Witness's AI output?: The source describes legal engineers comparing prompts with expected outputs and supplying source references in test environments. It does not name law-firm execution details. As a proposed design, the agent drafts and a qualified lawyer checks interpretation and advice, with no delegated sign-off.
How strong is the evidence for Orbital Witness's product?: Evidence is early-stage. The published account describes early law-firm deployments followed by a commercial product, so actual use is established. Named law-firm execution details, external tool boundaries and independently measured legal accuracy are absent.
What would an enterprise need to build something similar?: This is Nuvepro's proposed design, not Orbital's. It would reload an approved anonymized document bundle with a gold-standard issue set, include cases such as a missing annex or conflicting title record, and test issue recall, unsupported conclusions and escalation to qualified counsel.

Shape this around your enterprise

Bring a similar workflow and the people responsible for it. Together, define the records and systems to use, the result your team will build or configure, and how its owners will check it. Nuvepro scopes the simulation, environment and preparation. Agree any existing-agent evaluation, implementation assistance and production responsibilities separately.

Discuss this work Reimagine this workflow