Legal
Privacy
Agent One processes stakeholder interview data on behalf of the client that commissioned the engagement. This page states how that works.
Full document
The formal policy, reviewed against the Australian Privacy Principles and the Notifiable Data Breaches scheme, is in preparation. For a specific question about how your data is handled — or your organisation’s — email hello@1qlabs.ai and we will answer it directly.
What the platform holds
- Interview transcripts and messages from stakeholders who take part in an engagement.
- Documents stakeholders choose to upload for analysis.
- Derived process intelligence: blueprints, findings and proposals.
- Account records for the client staff and operators who access the platform.
Where it is held
- Inside a Google Cloud project and network isolated to that engagement.
- In Australian regions, enforced by organisation policy rather than configuration convention. Model inference is the exception: where a model is not served from an Australian region, that call is processed elsewhere in Google Cloud under the Cloud Data Processing Addendum.
- Never aggregated across clients, and never reachable from another engagement.
What is never done with it
- It is not used to train models. Vertex AI’s terms prohibit training on customer data.
- It is not sent to a model provider’s own API directly from a client cell.
- It is not sold, shared or used to enrich a third-party dataset.
Retention and deletion
- Retention follows the policy the client specifies at the start of an engagement.
- Export and deletion cover transcripts, embeddings, blueprints and findings.
- Deletion requests are actioned on the client’s instruction.