> For the complete documentation index, see [llms.txt](https://docs.dapta.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.dapta.ai/dapti/privacy-and-scope.md). # Privacy & Scope Dapti's job is to help you operate your Dapta account. To do that it needs access to your data, but only your data. This page explains exactly what Dapti can and cannot see, where the boundaries are, and how anonymized benchmarks are kept safe. *** ## What Dapti Can See When you ask Dapti a question, it can read everything inside your own workspace, including: * **Agents**: your voice and text agents, their prompts, voices, language settings, and post-call variables. * **Calls**: every call placed or received by your agents, with full transcripts, durations, outcomes, and sentiment. * **Meetings**: every meeting your AI Note Taker has joined, with transcripts, participants, and timestamps. * **Flows**: every Flow Studio automation you have built, with their nodes, configuration, and execution history. * **Scheduled tasks**: every cron job and recurring task in your workspace. * **Account context**: your plan, credit balance, usage history, connected integrations, and onboarding answers. * **CRM-style data**: contacts, lead lists, and campaigns that live inside Dapta. {% hint style="info" %} Dapti uses this access in real time, only when you ask. It does not run analyses in the background or pre-cache your data. {% endhint %} *** ## What Dapti Cannot See Dapti is scoped strictly to your workspace. It cannot: * **Read other companies' data**. Even when you ask about industry benchmarks, Dapti only sees aggregated averages, never any individual peer account. * **Access other users in your company without permission**. Dapti respects the same role and permission model as the rest of the Dapta platform. If a user does not have access to a flow, neither does Dapti when that user is logged in. * **Reach into your CRM, calendar, or email unless you have explicitly connected the integration**. When an integration is missing, Dapti will tell you and provide a link to connect it. It does not silently fall back to anything. * **Modify billing or your plan**. Dapti can answer questions about pricing and route you to the right page, but plan changes always happen through the in-app billing flow. * **Take destructive actions without confirmation**. Anything that modifies a live agent, deploys or changes a flow, or schedules a recurring task is always presented as a confirm-before-apply step. *** ## Customer Dapti vs. Internal DaptaIQ There are two assistants with a similar look and feel inside the Dapta platform: * **Dapti** (this product): workspace-scoped. Reads only your own data. Privacy-safe. Available to every Dapta customer. * **DaptaIQ**: an internal-only assistant used by Dapta employees for cross-account analytics. Not available to customers. If you ever see the name "DaptaIQ" anywhere, that is the internal product, not Dapti. *** ## How Industry Benchmarks Work When you ask Dapti how you compare to other companies in your industry, here is what actually happens: 1. Dapta computes industry-level statistics across all customers in a given industry (Real Estate & Construction, Dental, Ecommerce, etc.) using **aggregated, anonymized** data. 2. The result is a single set of averages per metric per industry. There is no way to trace any number back to a specific company. 3. Dapti shows your numbers next to those aggregates so you can see where you stand. You are part of those aggregates too, but only as one anonymous data point alongside many others. *** ## Audit and Transparency Dapti's reasoning is auditable. Every response includes: * A collapsed "**Thought for Ns**" indicator. Click it to open the Activity panel and see the step-by-step reasoning in plain English. * Source citations whenever Dapti pulls information from the web or from your data. * A clear note when Dapti cannot answer a question (for example, missing data, missing integration, or out-of-scope request) instead of guessing. *** ## Reporting a Concern If you ever see Dapti behave in a way that looks like a privacy issue (mentioning data that should not be visible, accessing something you did not connect, or returning information about another account), let us know: * Submit feedback through the **👎 thumbs-down** button on the offending Dapti response. * Contact us at **** for anything urgent. *** ## Next Steps {% content-ref url="/pages/XgNL8X3CGabWk1aDE5X4" %} [Tips & Prompting](/dapti/tips-and-prompting.md) {% endcontent-ref %} {% content-ref url="/pages/tGdZKRq1oCqpksmxm3rK" %} [Capabilities](/dapti/capabilities.md) {% endcontent-ref %} {% content-ref url="/pages/z2MqYMcg82nfTMVD8SSU" %} [Use Cases](/dapti/use-cases.md) {% endcontent-ref %} --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://docs.dapta.ai/dapti/privacy-and-scope.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.