# Interaction patterns

Source: https://developer.avalara.com/tax-transparency/zoh8986211306269/

# Interaction patterns

Learn about the 3 primary interaction patterns the AI agent uses to enhance tax transparency on transaction pages.

The AI agent uses 3 primary interaction patterns to support tax transparency within transaction pages. These patterns determine how insights and explanations are presented to users.

-   **Automatic insights**:

    When a transaction page loads, the AI agent automatically displays relevant tax insights within the UI. These insights appear without requiring user action and provide contextual information related to the transaction.

    Automatic insights may include certificate expiration warnings, detection of special tax scenarios, and postcalculation context indicators.

-   **On-demand explanation**:

    Users can request a detailed explanation by selecting the Explain the tax option. The AI agent retrieves the associated AvaTax transaction data and generates a comprehensive, plain-language breakdown of the tax calculation.

    The explanation may include jurisdiction-level details, applied tax rates, sourcing methodology, and any special rules that influenced the result.

-   **Conversational follow-up**:

    After presenting an explanation, the AI agent supports natural follow-up questions. Users can ask additional questions to clarify specific aspects of the tax calculation.

    For example, users ask why a particular jurisdiction rate applies. They want to know when a certificate expires. They also want to understand how a change in transaction details affects tax. The AI agent maintains conversation context to provide consistent and relevant responses.