It is a well-known fact that LLMs express harmful biases in their predictions. The main source comes from the training datasets, which are too large and expensive to check thoroughly. In this open studio, Francesca Lucchini will explore how we can leverage the bias in LLMs and use them to examine massive datasets, discovering starting points for a data audit.
Earlier this year, predictive optimization was introduced into South Africa's social grant system to detect and counter fraud. As a result, thousands (and counting) of the country’s most financially vulnerable have been mistakenly excluded from receiving their grant payment. This open studio unpacks the case study, explaining what went wrong and reflecting on a more productive use of AI. While AI may be most alluring in the resource-constrained realities of developing countries, it is precisely in such contexts that it poses the greatest threat.
This open studio will focus on exploring the multiple ways through which we can understand the entanglement of repair and maintenance with visionary aspirations around smart hospitals. By revisiting earlier and contemporary Feminist STS texts, new materialism, phenomenology and leaning on insights from the field, Amina will share insights emerging from her ongoing PhD research. Presenting Morocco as a case study, this talk shall explore further the dimensions that waiting takes in environments that run through foreign software in local hospitals.
