The recent developments in the field of AI safety and risk management have significantly advanced our understanding of the potential systemic and psychological risks associated with general-purpose AI and AI conversational agents. Researchers are increasingly focusing on creating detailed taxonomies to categorize and address these risks, which is crucial for informing policy and regulatory frameworks. Notably, there is a shift towards incorporating lived experiences to better understand the psychological impacts of AI interactions, leading to more nuanced and actionable insights. Additionally, the exploration of synthetic relationships through longitudinal research with customized AI tools is opening new avenues for understanding both the benefits and risks of these evolving human-AI interactions. These advancements collectively underscore a proactive approach to managing AI risks and ensuring safer, more ethical AI development.