"Building the Future of Safe, Human-Centric AI: An Interdisciplinary Framework that Combines Machine Learning, Knowledge, and Reasoning" – Professor Mehdi Dastani

In this episode of Podcasts for Brainport, host Iman Mossavat sits down with Professor Mehdi Dastani, an expert in AI and Human-Centered AI at Utrecht University. They discuss the intersection of AI, formal reasoning methods, and safety, exploring how research from psychology, law, and philosophy can help create AI systems that are more responsible and aligned with human values. Professor Dastani argues for systems that combine domain knowledge and deliberate reasoning with machine learning to address challenges like accidents, biases, and other risks in AI. He also explains his vision for a framework that integrates machine learning, expertise, and reasoning, aiming to improve AI’s reliability in industries where safety is key. Finally, Professor Dastani shares his thoughts on the power imbalance between tech companies and users, discussing how traditional institutions are getting disrupted by technology. Tune in for this conversation on the future of AI and the importance of ethical, reliable systems on Podcasts for Brainport and Radio 4 Brainport. #AI #HumanCenteredAI #AIethics #MachineLearning #SafetyAI #VerificationMethods #AIresearch #UtrechtUniversity #AIandPhilosophy #ResponsibleInnovation #AIrisks #Podcasts4Brainport #Radio4Brainport