Send us a text

Welcome to the fourth episode of our podcast series, where we delve into the intriguing world of AI and its impact on our society. In this episode, we explore the concept of Constitutional AI and the Uncertainty Principle. We examine the role of diverse cultures, values, and sensitivities in AI alignment, and the pressing question of who gets to decide critical aspects of alignment. We uncover the bias in data and design that has inadvertently oppressed certain groups, leading to real-life consequences. Additionally, we discuss the challenge of incorporating uncertainty into AI decision-making and explore innovative methods such as Stepwise Relative Reachability and Attainable Utility Preservation. Join us as we navigate these complex issues and strive for a future where AI is aligned with our shared values.

Website: synapticlabs.ai
Youtube: https://www.youtube.com/@synapticlabs
Substack: https://professorsynapse.substack.com/