In this episode, Xavier Bonilla has a discussion with Karl Friston about the Free Energy Principle, Bayesian Brain, and many neurobiological principles. They give an overview of the history of the models of the brain and discuss the evolutionary components of the brain. They also mention how important executive functioning and abstraction are for various organisms. They give an overview of the Bayesian brain, discuss predictive coding, and the two classes of neurons for prediction and prediction error. They discuss the free energy principle, Markov blanket, and homeostasis. They discuss consciousness and the role of active inference, artificial intelligence, and various neuroimaging techniques. They mention how the free energy principle could be used pragmatically and many other topics. 
Karl Friston is a theoretical neuroscientist at University College, London. He is a Fellow of the Royal Society in Biology. He has been awarded numerous awards for his contributions to neuroimaging. Those contributions include the creation of Statistical Parametric Mapping (SPM), Voxel-Based Morphometry (VBM), and Dynamic Causal Modeling (DCM). These contributions were heavily used in helping clinicians understand individuals with Schizophrenia and other disorders. You can find his published research here.