AbstractAutomated scientific discovery is a discipline which lies at the boarder of artificial intelligence, natural sciences and philosophy of science, and deals with the application of artificial intelligence methods to scientific discovery. Historically, its origins go back to the 1960s and there have been at least three major research programs in the field, each of them having different objectives, concerns, and methodology: machine learning systems in the Turing tradition, normative theory of scientific discovery formulated by Herbert Simon's group, and the programs called HHNT, proposed by J. Holland, K. Holyoak, R. Nisbett, and P. Thagard. In the paper I briefly describe new developments in the field, recent issues in machine learning and data science applications to scientific discovery, and explore lessons for the philosophy of science that can be drawn.