The field of artificial intelligence is moving towards a greater emphasis on safety and literacy, particularly in regards to the protection of young users and the development of more transparent and trustworthy AI systems. Recent research has highlighted the importance of evaluating and improving the ability of AI models to detect and mitigate harmful interactions, as well as the need for more effective content moderation practices on social media platforms. Additionally, there is a growing focus on fostering AI literacy in children, with studies exploring their mental models of AI reasoning and developing tools to help them critically engage with generative AI. Noteworthy papers in this area include: Selective Code Generation for Functional Guarantees, which proposes a method for controlling code hallucination in AI-generated code, Children's Mental Models of AI Reasoning, which identifies three models of AI reasoning in children and highlights implications for AI literacy education, and AI Puzzlers, which develops an interactive system to help children identify and analyze errors in generative AI.