Ok, but we can we agree we can definitely not trust OpenAI like 10x more?