Carmen researches the human layers of responsible AI, where technical systems meet cultural interpretation, organizational trust, and individual judgment. Her work spans GenAI safety and evaluation, AI governance, and the cross-cultural alignment challenges of AI systems deployed across diverse societies.
Originally from Hong Kong and based in Europe, her research spans biometric AI profiling risks (MIT Science Policy Review), LLM-robot governance (ACM CHI 2026), and AI policy analysis covering 80+ nations, including the EU, China, Japan, and the US (Center for AI & Digital Policy). As a research sprint fellow at Harvard's Berkman Klein Center for Internet & Society, she co-developed open educational frameworks on global AI governance. She holds an M.Sc. in Responsible Data Science from the Technical University of Munich, focusing on quantitative model auditing for factuality, explainability, and alignment gaps across languages. Before and alongside her research, she built over a decade of cross-sector experience at Siemens and Reuters News, covering sustainability technologies across global markets and breaking policy stories across the Asia-Pacific region in three languages, grounding her in the operational realities that responsible AI must navigate across industries, cultures, and functions.