Ontogenetic Epistemology of AI: An Investigation into the Role of Machine Learning in Knowledge Production Through its Genesis and Genealogy

Through a reconstruction of Gilbert Simondon's work using perspectives from contemporary philosophy of science, I pursue an ontogenetic study of machine learning —an investigation into the process by which machine learning models become what they are and do what they do —with the aim of understanding and addressing their role in knowledge production. This investigation unfolds as two parts: Genesis and Genealogy. By studying the genesis of the ML-model, I seek to identify the role (or lack thereof) of human cognition in the formation of the ML-model and illustrate its significance for our knowledge about the world. By studying the genealogy of the ML-model, I aim to address how different scientific communities and their respective knowledge practices are affected by their interaction with and integration of machine learning models.