How do people living in rural villages in Togo feel about the use of emerging technologies in humanitarian aid? This work reports on the privacy concerns of people living in rural Togo related to the use of machine learning models trained on phone data to allocate cash assistance to people living in poverty, highlighting an innovative method -- sociotechnical visuals -- to explain complex technical concepts so that people living in rural villages with limited literacy, formal education, and familiarity with digital tech could provide meaningful input.

This Open Studio shares an in-progress Feminist AI–informed learning design framework that examines how power, authorship, and agency shift when Generative AI is introduced into secondary education. Drawing on classroom pilots, design research, and assessment tools such as a Student-Author Voice rubric, the work explores how LLMs can be integrated through co-creation and gradual immersion, rather than extraction or automation. The session foregrounds methodological tensions, ethical trade-offs, and policy-relevant questions emerging from real educational contexts.

Public sector AI has the potential to harm citizens, with risks increasing as its use expands. Recent work positions public procurement as a way to shape public sector AI in line with public interests, using the state’s purchasing power to influence which AI systems are procured and under what conditions. In this presentation, I explore how this potential could be realised in practice by drawing on semi-structured interviews with UK/EU buyers, providers, and procurement experts.