Artificial intelligence (AI) is reshaping scientific practices, yet its epistemic implications remain underanalyzed. While recent advances in large language models are substantial, machine learning (ML) has a deeper history across disciplines. This manuscript examines 4.9 million publications and 255 ML techniques to understand how the latter are reconfiguring scientific methods and knowledge production. Through embedding-based mapping, we reconstructed the semantic space of ML research, and found a core-periphery structure where physical sciences form the methodological core and health sciences represent the primary area of adoption. Methodological profiles vary by domain: predictive techniques are concentrated in computer sciences, while inferential approaches remain distributed across applied fields. Predictive architectures, however, are displacing inference-oriented techniques in domains that have traditionally prioritized interpretability, such as health and social sciences. This displacement unfolds in two distinct waves: first (2015-2021), through deep learning architectures that reduced predictive error at the expense of epistemic opacity; and second (post-2022), through reliance on external commercial models that introduce further layers of opacity over inaccessible data and processes. This transformation expands science's analytical capacity and reshapes how knowledge is produced and evaluated.