This podcast episode delves into the growing concerns around AI risk, highlighting issues such as bias, potential harm, and deception. It emphasizes the need for a new technical approach and governance system to mitigate these risks. While acknowledging the tension between AI's societal benefits and corporate interests, the episode advocates for a global, non-profit, and neutral international agency to oversee AI development. It also underscores the urgency of global governance, given the limitations and risks posed by current models like ChatGPT.