Abstract
: This paper examines the ethical foundations that guide the responsible creation and deployment of Machine Learning (ML). Given how rapidly ML is gaining influence in healthcare, finance, and public policy, it is
increasingly vital to uphold applications that promote transparency and societal benefit. We highlight ten core principles—accuracy, bias, accessibility, security, privacy, transparency, accountability, human oversight, sustainability, and harm avoidance—and illustrate ways to implement them so that ML systems strengthen social well-being
rather than undermine it. Drawing on theoretical perspectives alongside real-world illustrations, we outline best
practices that foster trust and responsible progress in ML. Ultimately, we argue that robust governance structures
guided by these principles will help steer ML-based projects to become genuine engines for positive social change.