The link below is the question from the customer about auto-scaling. It is not clear how this exactly works in ROSA.
https://access.redhat.com/documentation/en-us/red_hat_openshift_service_on_aws/4/html/cluster_administration/nodes?extIdCarryOver=true&sc_cid=7013a000002pdO4AAI#rosa-nodes-about-autoscaling-nodes 

Per the docs, it says, “automatically scale the number of machines in a cluster.” so, in a multi-az cluster, there will be 3 machine sets.

We need more explanation for this.
Here are what I learned from engineering, and I provided the following to the customer.

This is the reference to how the K8S cluster auto-scaler works: https://github.com/kubernetes/autoscaler/blob/master/cluster-autoscaler/FAQ.md#im-running-cluster-with-nodes-in-[…]t-supported-by-cluster-autoscaler. I referenced the question for your use case.
ROSA uses --balance-similar-node-groups flag as true in the default setting 
$ oc get clusterautoscalers.autoscaling.openshift.io default -o yaml
apiVersion: autoscaling.openshift.io/v1
kind: ClusterAutoscaler
metadata:
  creationTimestamp: "2022-12-22T21:37:43Z"
  generation: 1
  name: default
  resourceVersion: "187750893"
  uid: c2f78f86-2c9e-43cb-a1af-d0a71987532d
spec:
  balanceSimilarNodeGroups: true
  scaleDown:
    enabled: true
Cluster Autoscaler will only add as many nodes as required to run all existing pods. If the number of nodes is not divisible by the number of balanced node groups, some groups will get 1 more node than others.
The Kubernetes/autoscaler reference also mentions that similar node groups will not have the exact sizes.

1.
2.
3.

The customer believed that if additional resources were required and autoscaling enabled, the nodes would scale in a set of three since they have multi-az for the cluster per how the docs describe it currently. 

if we can provide the details of how the cluster auto-scaler works and descript the expected ROSA behavior.