Last updated: May 2026
This policy explains how OpsPilot uses artificial intelligence, what data is sent to AI models, and what safeguards are in place to protect your business and your clients.
Core Principle: AI Suggests, You Decide
OpsPilot never takes autonomous actions on your behalf. Every briefing, email draft, and recommendation requires your explicit review and approval before anything happens.
When you generate a briefing or email draft, OpsPilot sends the following data to the AI model:
| Data Type | What Is Sent |
|---|---|
| Business Context | Agency type, services, client types, AI priorities (from your AI Memory) |
| Communication Style | Your preferred tone, style notes, and company rules |
| Briefing Input | Text you paste into the briefing form (email summaries, meeting notes, client updates) |
| Client Names | Client names you have added to OpsPilot (not email addresses or contact details) |
Important: OpsPilot does not automatically send your raw email content to the AI. When Gmail integration is available, only email metadata and summaries will be processed — never full email bodies without your explicit action.
OpsPilot uses a large language model (LLM) provided through a secure API. The model processes your requests in real time and does not retain your data between sessions. We do not use your data to fine-tune or train AI models.
To ensure fair usage and prevent abuse, AI generation is rate-limited to 20 requests per hour per user. If you reach this limit, you will be notified and can try again after the window resets.
If you have concerns about how AI is used in OpsPilot, contact us at [email protected].