In this talk, the author describes novel approaches, models, and case studies in the Bayesian Brain (AI) systems, based on knowledge and expertise coming from diverse but corresponding disciplines. It is assumed that the human brain represents the statistical structure of the world at different levels of abstraction by maintaining different causal models that are organized on different stages of a hierarchy, where each stage obtains input from its secondary stage. Author proposes a dynamic method of communication between bots and humans in order to Intent generations (an intent is the desire to complete an action). The focus of most algorithms for Intent generation is to create a non-ambiguous description, but this is not how people naturally communicate. The authors call dynamic description how humans tend to give an underspecified description and then rely on a strategy of repair to reduce the number of possible actions or objects until the correct one is identified.