The field of artificial intelligence is rapidly advancing, with a growing focus on responsible AI development and deployment. Recent research has emphasized the importance of transparency, accountability, and ethics in AI systems, particularly in high-stakes applications such as healthcare and security. A common theme among recent studies is the need for more nuanced and context-dependent approaches to AI governance, moving beyond simplistic explanations and towards more comprehensive frameworks that incorporate multiple stakeholders and perspectives. For instance, the proposal of a pro-justice EU AI Act toolkit aims to provide a comprehensive framework for AI ethics and governance, while the development of a unified framework for human-AI collaboration in security operations centers integrates AI autonomy, trust calibration, and human-in-the-loop decision making. The field of generative AI is also rapidly evolving, with a growing focus on mitigating risks and ensuring responsible development. Researchers are exploring the potential threats posed by open-weight AI models, including accelerated malware development and enhanced social engineering. To address these concerns, there is a need for pragmatic policy interpretations, defensive AI innovation, and international collaboration on standards and cyber threat intelligence sharing. Another critical sustainability challenge is the issue of digital waste, or stored data that consumes resources without serving a specific purpose. Studies are investigating the impact of generative AI on administrative burdens, trust dynamics, and the effectiveness of labeling AI-generated images in reducing misinformation. Noteworthy papers include Mitigating Cyber Risk in the Age of Open-Weight LLMs, which proposes evaluating and controlling specific high-risk capabilities rather than entire models, and Responsible Data Stewardship, which introduces digital waste as an ethical imperative within AI development and proposes strategies to mitigate its environmental consequences. In the field of public sector technology, researchers are working to develop systems that are grounded in the goals and organizational contexts of employees, with a focus on surfacing important equity considerations around data and technology use. This includes exploring the challenges that employees face when operationalizing equity, as well as the design space for acceptable government technology. The integration of research data management into teacher training programs is also being explored, with a focus on adequately preparing future teachers to handle data competently. Noteworthy papers in this area include How Soft Skills Shape First-Year Success in Higher Education, which presents a teaching intervention aimed at fostering key soft skills among first-year computer science students, and shows significant improvement in individual presentations and team projects. Finally, researchers are focusing on developing more safe and trustworthy models, with a growing emphasis on transparency and accountability in dataset documentation. New benchmarks and evaluation metrics are being proposed to assess the sensitivity of models to dual-implicit toxicity, and general-purpose generation models are being developed to unify diverse tasks across modalities within a single system. Noteworthy papers include MDIT-Bench, which introduces a novel toxicity benchmark for evaluating dual-implicit toxicity in large multimodal models, and ComfyMind, which presents a collaborative AI system for general-purpose generation via tree-based planning and reactive feedback. Overall, these advances have significant implications for the future of AI research and governance, and highlight the need for ongoing investment in this area to ensure that AI systems are developed and deployed in ways that prioritize human well-being and safety.