Cheng​ Yu​

Research Scientist​

Cheng Yu is a master’s student in the Risk and Safety program at the Technical University of Munich (TUM), affiliated with the Civic Machines Lab. Her research explores the safety implications of generative AI, with a focus on retrieval-augmented generation (RAG) and the emergence of bias and harmfulness in AI agents during web interactions. She is particularly interested in how access to external knowledge sources may influence AI alignment, and in developing evaluation frameworks for responsible AI behavior in real-world deployments.​ Drawing on her interdisciplinary training, Cheng aims to contribute to the development of AI systems that are not only powerful but also transparent, robust, and ethically aligned.​

Related Projects

The Civic Machines Lab explores the full lifecycle of artificial intelligence, from the semiconductors powering computation to the algorithms shaping human experience.
Situated within the TUM Think Tank, the lab views AI as a socio-technical system rather than a purely technical challenge.
By integrating perspectives from computer science, social sciences, and public policy, the Civic Machines Lab aims to shape technologies that are democratic, sustainable, and globally inclusive.

Lab

Scaling innovation across NGOs

In collaboration with WWF's Global Innovation Team and the Innovation for Impact Network (IFIN), the Civic Machines Lab is developing a validated framework and AI-powered coaching tool to help NGO project teams design, scale, and measure innovation more effectively.

Project