WebRTC Recording Export Metadata Inconsistency in Legal Discovery Jobs

Just noticed that the recording export jobs initiated via the Recording API for WebRTC softphone interactions are failing to include specific metadata fields required for our legal discovery chain of custody. The environment is Genesys Cloud EU (eu-1) and we are using the standard bulk export data action within Architect flows to trigger these exports to our BYOC S3 buckets. The issue appears specifically when the interaction involves WebRTC channels where the recording is generated from the browser-based softphone rather than traditional telephony. When reviewing the exported JSON manifests, the interaction_id and recording_url are present, but critical fields such as participant_external_contact_id and web_rtc_session_id are missing or nullified, even though they are clearly visible in the interaction detail view within the Admin console. This is causing significant delays in our legal hold processing as we cannot properly link the audio files to the specific digital channel sessions for audit trails. The Create Recording Export Job returns a 200 OK status, indicating the job was accepted, but the subsequent failure in metadata inclusion suggests a filtering issue or a known limitation with WebRTC recording metadata propagation. We have verified that the user roles have the necessary permissions for recording export and interaction viewing. Additionally, the S3 bucket policies are correctly configured to allow writes from the Genesys Cloud service role. The problem is consistent across multiple export jobs triggered over the past week, affecting both individual recordings and batch exports. Has anyone encountered similar issues with WebRTC recording metadata in export jobs? We are looking for a workaround or a confirmation if this is a platform limitation that requires a specific configuration change in the Architect flow or API payload. The current workaround involves manually extracting metadata from the interaction API, which is not scalable for large legal discovery requests involving thousands of digital channel interactions. Any insights or documentation references regarding WebRTC recording metadata handling in export jobs would be greatly appreciated.