What is the importance of monitoring the performance of a customer service chatbot after deployment?

A$.$ Because of the LLM$\text{'}$s knowledge cutoff, we must continuously monitor the knowledge cutoff and update its knowledge frequently.

B$.$ Every product should be monitored to track customer satisfaction $--$ this is good practice for all software.

C$.$ This is false. So long as internal evaluation is done well, further monitoring is not necessary.

D$.$ In case customers say something that causes the chatbot to respond in an unexpected way, monitoring lets you discover problems and fix them.