The paper aims to present an overview of previous research on “machine learning” applications in banking, covering key aspects of recent discoveries, their limits, and potential future research directions. It makes two contributions to the body of knowledge. It initially divides the literature to provide an overview of completed research endeavors. Second, it points out a gap in the existing body of research and suggests fresh avenues for investigation. The findings indicate that prior research has had difficulty developing a sound theoretical foundation for the subject. To support the proposed "theories," "notions," and "paradigms," more study is needed. In short, there is a big need for more research because there hasn't been a thorough evaluation of how machine learning has been used in banking.