There is no error code in the answers that are provided by AI prompts. It will return an answer that is the best fit to the prompt or the question, but it does not tell you the probability that it is not correct or that it is low probability of answer. The conversational AI will not return a probability function though it does follow a reasoning model. An easy example is to ask your AI model to create a picture of you based on your resume or based on your characteristics that you describe. It will create a face that is a representation of you but it is not you and it cannot create the exact you. That is expected. However what is unusual is that it does not give you an error on what it created nor does it actually tell you that it is not true.
Observability is important and human in the loop may be critical.
