At best they'll downplay the mistakes and assert that it's a problem that'll fix itself despite the fact that OpenAI admitted that it *won't*.