In this episode, Somy and JK talk to Sam and Josh, the founders of Decoded AI, a research lab that focuses on making AI risk identifiable, quantifiable, and valuable. They discuss topics such as how to define and measure functional safety for AI systems, how to identify and mitigate common AI security risks such as data poisoning and adversarial attacks, how to substantiate claims about AI systems to build trust and persuade users to seed decisions to autonomous systems, how to align AI systems with human values and ethical principles, how to communicate effectively with stakeholders and users about AI systems, and how to keep up with the rapid innovation and predict the future of AI. This podcast is a great resource for anyone who wants to learn more about the challenges and opportunities of AI security and ethics.