Legal document review · Production

Orbital Copilot AI Agent accelerates real estate lawyers' lease reporting by up to 70%

The problem

Real estate lawyers must wade through hundreds of pages of intricate legal documents to perform due diligence, a highly time-intensive process. Classical ML approaches required extensive labelled datasets, and simpler LLM-based systems were limited to straightforward tasks and unable to replicate the complex, multi-document reasoning lawyers apply during due diligence.

First attempt

Simple LLM wrapper systems that summarise documents or answer specific one-off questions were insufficient for the complex, multi-document reasoning required in real estate legal due diligence.

Workflow diagram · grounded in source
1
Lawyer submits query and documents
trigger
“When prompted with the query "What is the rent and how has it been varied?", Orbital Copilot leaps into action. It begins by understanding the question's context and the documents at hand.”
2
Agent searches document sections
ai_action
“it meticulously searches and reads the pertinent sections in both documents”
3
Supplementary legal research
integration
“It seeks out additional legal information to refine its understanding of the lawyer's initial instructions. This could be a proprietary legal knowledge base or data from HM Land Registry”
4
Agent analyses and formulates response
ai_action
“Orbital Copilot analyses its findings, formulates a response, and presents it for review”
5
Lawyer reviews with cited sources
human_review
“Orbital Copilot indicates the parts of the PDF documents it consulted, facilitating direct navigation to the primary evidence supporting its answers so lawyers can immediately see and trust where an answer came from”
Reported outcome

Orbital Copilot reduces the time for a comprehensive lease report by up to 70%, yielding substantial financial savings per property for law firms and their clients, with cumulative efficiency gains given the regularity of lease reports in real estate law.

Reported metrics
Lease report time reductionup to 70%
Lease report baseline duration per property2-10+ hours depending on complexity
Financial savings per propertysubstantial financial savings
Cumulative efficiency and cost savingssubstantial
Reported stack
Orbital CopilotGPT-4BERTT5HM Land Registry
Source
https://tech.orbitalwitness.com/posts/2024-01-10-we-built-an-ai-agent-that-thinks-like-a-real-estate-lawyer/
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Orbital Copilot reduces the time for a comprehensive lease report by up to 70%, yielding substantial financial savings per property for law firms and their clients, with cumulative efficiency gains given the regularit…

What tools did this team use?

Orbital Copilot, GPT-4, BERT, T5, HM Land Registry.

What results were reported?

Lease report time reduction: up to 70%; Lease report baseline duration per property: 2-10+ hours depending on complexity; Financial savings per property: substantial financial savings; Cumulative efficiency and cost savings: substantial (source-reported, not independently verified).

What failed first in this deployment?

Simple LLM wrapper systems that summarise documents or answer specific one-off questions were insufficient for the complex, multi-document reasoning required in real estate legal due diligence.

How is this legal document review AI workflow structured?

Lawyer submits query and documents → Agent searches document sections → Supplementary legal research → Agent analyses and formulates response → Lawyer reviews with cited sources.

Related legal document review cases