Why does this setting… regarding the inclusion of internal bot logic steps in the exported transcript JSON appear to be ignored when using the bulk export API for digital channels?
We are currently running Genesys Cloud v2024.07.1. The requirement for our legal discovery process is to maintain a strict chain of custody for all customer interactions, including the AI bot’s internal decision-making path. We are initiating exports via POST /api/v2/analytics/conversations/export with the filter set to include ‘bot’ interactions. The manifest.json is generated correctly, and the audio files for voice channels are present.
However, when reviewing the transcript.json for Web Chat sessions handled by the AI Bot, the ‘internal’ nodes from the Architect flow are missing. Only the final text sent to the user is captured. The API documentation suggests that the ‘include_internal_steps’ flag should populate these, but the field remains null in the export payload. This creates a gap in the audit trail, as we cannot verify why the bot escalated to a human agent. We have verified the flow configuration and the export job settings multiple times. The issue persists across different bot versions. Is this a known limitation of the bulk export API for digital channels, or is there a specific configuration in the Architect flow that enables this metadata capture for legal exports?