Watch a 46-minute lecture from Frauke Kreuter of LMU Munich and University of Maryland exploring the critical challenges of aligning AI systems with human values and norms. Examine historical approaches to measuring public preferences over the past 70 years, including common errors and pitfalls in capturing true human intentions. Learn about systematic biases, misinterpretations, and flawed instrument design that can impact human feedback loops in AI development. Discover valuable benchmark databases for aligning AI with human values while considering the dynamic nature of societal norms across different cultural contexts. Gain insights into collaborative opportunities between computer science and social science researchers to improve the measurement and integration of human norms into AI models.