Last updated September 2026. This describes how Op by Op handles your data. For contractual processing terms, see our Data Processing Agreement.
An uploaded procedure is used only to generate your analysis, then discarded. It is never added to your saved record — only the structured analysis it produces is stored. If an analysis fails, the file is retained briefly so you can retry, then automatically purged. We do not build a library of your source documents.
Analyses are generated on enterprise-grade AI infrastructure. Your data is never used to train AI models and is never shared with model providers. Prompts and outputs are not retained by the model provider for training or any other purpose.
Each organization's data is isolated from every other organization, enforced at the database layer. Access within your organization is controlled by role (owner / member). We access your content only as needed to operate and support the service, or as you direct.
Every change to an analysis is recorded in an append-only, tamper-evident audit log (a per- organization cryptographic hash chain), so the history of your process knowledge is verifiable.
Your data is encrypted in transit (HTTPS enforced) and at rest. The service runs on SOC 2 Type II cloud infrastructure with access controls and per-organization isolation. We can provide further security detail under a mutual NDA as part of your review.
Your analyses and account data are retained while your account is active. You can request export or deletion of your organization's data at any time; on account closure we delete it within a commercially reasonable period.
We use a small number of subprocessors by category — cloud hosting and enterprise AI infrastructure — each bound to protect your data and not to use it for their own purposes. A current list of named subprocessors is available on request.
Questions, export/deletion requests, or a subprocessor list: privacy@opbyop.com.