Databricks cluster usage permissions

Hi,
I am trying to manage Azure databricks cluster usage using databricks_permissions resource:

resource "databricks_permissions" "cluster_usage" {
  cluster_id = databricks_cluster.shared_autoscaling.id

  access_control {
    group_name       = "aad_group_name"
    permission_level = "CAN_ATTACH_TO"
  }
}

we’re following Microsoft’s best practices and we’re using only account level groups. We have enabled identity federation for all our workspaces too. However, when trying to deploy (terraform apply) resource block from above, terraform returns an error:

Error: cannot update permissions: Principal: GroupName(aad_group_name) does not exist

This group exists in databricks, at account level though. So I created this group “aad_group_name” in databricks on workspace level, and that worked fine.
My question is, when using databricks_permissions resource, am I limited to use workspace-level group(s) only?
All the documentation about this resource seems to use workspace level groups, but then it conflicts with Microsoft recommendation to use account-level groups in Databricks.

2 year late, I ran into this issue as well.

I was able to go into Databricks, and add the account level groups to the workspace manually. Then Terraform / Databricks provider was able to see the group for the permissions.

Not sure if there is a way to have terraform add the group to the workspace after creating it.