diff --git a/factory/aws/src/aws.rs b/factory/aws/src/aws.rs
index ab8011a..40834d0 100644
--- a/factory/aws/src/aws.rs
+++ b/factory/aws/src/aws.rs
@@ -94,21 +94,28 @@ async fn destroy_instance(
     id: &str,
     force_stop: bool,
 ) -> Result<()> {
-    /*
-     * Before terminating an instance, attempt to initiate a forced shutdown.
-     * There is regrettably no force flag for termination (!) and if we don't do
-     * this first, AWS will sometimes wait rather a long (and billable) time
-     * before actually terminating a guest.  It is difficult, as the saying
-     * goes, to get a man to understand something, when his salary depends upon
-     * his not understanding it.
-     */
-    if force_stop {
-        info!(log, "forcing stop of instance {id}...");
-        ec2.stop_instances().instance_ids(id).force(true).send().await?;
-    }
-
     info!(log, "terminating instance {id}...");
-    ec2.terminate_instances().instance_ids(id).send().await?;
+    ec2.terminate_instances()
+        .instance_ids(id)
+        /*
+         * If requested, perform a forced shutdown of the VM before terminating
+         * it, otherwise AWS waits until the instance stops on its own (still
+         * billing you for it!).  Note that both `force` and `skip_os_shutdown`
+         * must be enabled for the instance to reliably shut down immediately:
+         *
+         * - `force` ensure the instance gets terminated if there is a problem
+         *   with the underlying compute host.
+         *
+         * - `skip_os_shutdown` avoids sending an ACPI shutdown signal to the VM
+         *   and immediately stops it, instead of waiting for the instance to
+         *   shut itself down.
+         *
+         * https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/instance-stop-methods.html
+         */
+        .set_force(Some(force_stop))
+        .set_skip_os_shutdown(Some(force_stop))
+        .send()
+        .await?;
 
     Ok(())
 }