Initially I also thought that this could be a bug, but as can be seen in screenshot, the keypad is numeric, and alphabets can’t be typed, so manually its not reproducible. But only when script is trying to type its able to.
Ok, that’s reasonable. Ever heard the joke about the patient who tells the doctor, “It hurts when I do this.”. Doctor says, “Don’t do that.”. I think you are in a situation where the automation is going beyond the user’s ability when they use the app. You may have to just rework the test case.
You can, of course, file a bug if you are so inclined. Seems like one to me.