# AI

The **AI** section in Settings allows administrators to manage **AI providers**, define **data-sharing policies**, and control how contextual information is shared with AI models during analysis, conversations, and agent executions.

<figure><img src="/files/y7kcNuV0Lb51V015cpIy" alt=""><figcaption></figcaption></figure>

## Data Sharing Permissions

At the top of the page, you can configure **what type of data is allowed to be shared** with AI providers. These settings apply globally and define the privacy and governance boundaries of AI usage inside Gaio DataOS.

#### 1. Metadata

Basic structural information about tables is shared.

**Included**

* Column name
* Column title
* Data type

**Excluded**

* Sensitive data values

**Purpose**\
Helps the AI understand schema structure without accessing actual data.

#### 2. Table Sample Data

A limited subset of table data is shared.

**Included**

* First 10 rows
* Sensitive data (if enabled)

**Purpose**\
Improves contextual understanding for more accurate responses in conversations involving data interpretation.

#### 3. Analytics

Derived analytical information from tables.

**Included**

* Frequency distributions
* Metrics
* Analytical summaries
* Sensitive data (if enabled)

**Purpose**\
Allows the AI to reason over patterns, trends, and aggregates rather than raw data only.

#### 4. Results Sample

A subset of query execution results is shared.

**Included**

* First 10 query results
* Sensitive data (if enabled)

**Purpose**\
Provides execution context for follow-up questions and complex analytical reasoning.

***

## Creating or Editing an AI Configuration

1. Navigate to **Settings → AI**
2. Click **New** or **Edit** on an existing configuration
3. Fill in the following fields:

#### Fields Description

* **Name**\
  Internal identifier for the AI configuration.
* **Supplier**\
  Select the AI provider (e.g., OpenAI, Anthropic, DeepSeek).
* **Model**\
  Choose the specific model available for the selected supplier (e.g., GPT-4, Claude Opus, DeepSeek Chat).
* **API Key**\
  Paste the provider-issued API key. This key is stored securely and used at runtime.

4. Click **Save**

***

### Preferential Model

You can mark one AI configuration as **Preferential**.

* The preferential model is used by default across:
  * Dashboards
  * Agents
  * AI-powered features
* Other models remain available for manual selection or advanced use cases.

### Governance and Security Notes

* Sensitive data sharing is **explicitly controlled** per data scope.
* API keys are never exposed to end users.
* AI suppliers only receive the data scopes you explicitly enable.
* You can mix providers (e.g., OpenAI + Anthropic) in the same environment.

### Best Practices

* Start with **Metadata only**, then progressively enable sample data if needed.
* Use **Table Sample Data** for exploratory analysis and schema discovery.
* Enable **Analytics** only when deeper statistical reasoning is required.
* Define one **Preferential model** to keep behavior consistent.
* Use different providers to compare quality, cost, and performance.

{% hint style="info" %}
This configuration layer ensures Gaio’s AI capabilities remain **powerful, transparent, and compliant**, while giving teams full control over how intelligence is applied to their data.
{% endhint %}


---

# Agent Instructions: 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:

```
GET https://docs.gaiodataos.com/settings/ai.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
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.