Working with "CredentialUIBrowser.exe"

A trick that might work is to put the activity inside a try/catch. Then in the exception block put a Message Box activity. This will stop the automation and make it sit waiting for someone to click OK.

When it reaches this point you can then log into the VM using the same robot account, and should be able to see what’s on the screen as well as use UI Explorer to verify the properties/selectors is the same.