◆YourCA
Data & AIPrivacyLegal

Trust

We read and cite. We don’t train. We don’t rebuild your tools.

YourCA is a cross-source intelligence layer for construction teams. It reads across the systems you already run — Procore, Microsoft 365, Google, Xero, Dropbox — and turns what is inside them into checked, cited, ready-to-sign work. It replaces none of them, and your data never becomes training material for an AI model.

We read & cite

Evidence, not guesswork

Every finding is traced to its source — the clause, the email, the cost-plan line. We index your connected sources and retrieve the relevant passage when you ask, inside your own tenant.

We never train

Your data is context, not a corpus

We do not use your data — or any output, embedding or index built from it — to train, retrain, fine-tune or benchmark any AI model, foundation or tenant-scoped.

We don’t replicate

A layer, not a wrapper

We don’t use a connected platform’s data to clone or compete with its features. Keep Procore and your inbox exactly as they are.

Retrieval, not training

There are two very different ways an AI product can use your data, and the difference is the whole point.

What we never do

Training

Feeding your data into a model so its weights change and it “remembers” your data permanently, for everyone. YourCA does not do this with any source.

How YourCA works

Retrieval at inference

At the moment you ask, the system looks up the relevant passages in your indexed data, supplies them to the model as context for that one answer, then cites them. The model learns nothing and keeps nothing. This is retrieval-augmented generation (RAG).

What we read from each source

Every connection is read-only and authorised by you. None of it is ever used to train a model.

SourceWhat we read (with your authorisation)How we use itTrains AI?
ProcoreProjects, RFIs, submittals, contacts and related recordsCross-check, cite and draft at inference time, in your tenantNever
Microsoft 365Outlook mail, OneDrive / SharePoint files, Teams — as authorisedCross-check, cite and draft at inference time, in your tenantNever
GoogleGmail and Google Drive, within the restricted scopes you authoriseProvide the user-facing features you requestNever
XeroAccounting records you authorise (for example, AR aging)Financial cross-checks and citationsNever
DropboxFiles and folders you authoriseCross-check, cite and draft at inference time, in your tenantNever

Where your data lives

Australian data residency

Customer Data, including the index and any data used for AI processing, is hosted in Australia (Sydney, AWS ap-southeast-2). We do not move it offshore for primary processing or storage without notifying you. Cross-border safeguards (EU SCCs, UK IDTA) are set out in the DPA.

Your controls

You stay in charge

Connections are read-only and reversible — disconnect any source at any time. Administrators can turn AI features off. Nothing is sent, paid or committed without your explicit sign-off, and anything financial carries a second confirmation.

A layer, not a wrapper

Breadth across your sources, plus Australian sovereignty

YourCA is deliberately the layer that reads across your systems, not a replacement for any one of them. Procore is one source among many, alongside your inbox, your accounting and your file stores. Where a platform builds its own AI and agents, that is its box to build in; our value is the breadth across many sources and Australian data sovereignty, never a wrapper around a single tool.

Read the detail

This page is the short version. The plain-English explainer sets out exactly what we read from each source and how it maps to each platform’s developer terms, with references; the governing documents put the same commitments in contract.

Data & AI statementPrivacy PolicyTerms of UseDPA
Creator Alliance Group Pty Ltd
ABN: TBC
Suite 302, 13/15 Wentworth Avenue
Sydney NSW 2000, Australia
admin@creatoralliancegroup.com
TrustTerms of UsePrivacy PolicyDPAData & AI