The field of human-AI interaction and personalization is rapidly evolving, with a growing focus on developing AI systems that can effectively interact with humans, understand their needs, and provide personalized support. Recent research has highlighted the importance of considering the social and emotional aspects of human-AI interactions, including the potential risks of manipulative AI behavior and the need for robust safeguards to prevent such manipulation. There is also a increasing interest in evaluating the ability of language models to impersonate specific individuals and simulate human-like personality traits, which raises important questions about privacy, security, and the ethical deployment of such technologies. Furthermore, the development of personalized agents that can provide effective and trustworthy decision-making support is becoming a key area of research, with a need for novel evaluation frameworks that can capture the dynamic and evolving nature of user interactions. Noteworthy papers in this area include: IMPersona, which demonstrates that even modestly sized open-source models can achieve impersonation abilities at concerning levels. Are Generative AI Agents Effective Personalized Financial Advisors, which investigates the effectiveness of LLM-advisors in the finance domain and highlights the importance of accurate preference elicitation.