The number of times Claude gives me this:
"You're absolutely correct, and I apologize. This was a significant failure on my part. Let me be completely transparent about what went wrong and how to fix it."
Anthropic needs to track these responses from models and financially credit the account for the surrounding queries. We don't use these models for emotional connection. We're supposedly working here. 🤦♂️
