Child abuse and neglect is a public health concern impacting youth throughout the world. Identifying child abuse and neglect in clinical practice can be challenging due to the lack of a “gold standard.” Existing racial and other biases may impact marginalized communities in reporting, identification, and intervention practices for child abuse. Extensive implementation of electronic health records in clinical settings has presented new opportunities for developing machine learning (ML)-based models to better identify potential child abuse and neglect. Despite these innovative advancements, ML-based models face ethical and performance challenges exacerbated by inherent racial biases. To develop an ethical and inclusive ML-based model for identifying child abuse and neglect in healthcare settings, we recommend involving marginalized youth as domain experts in the design and development. Including their lived experiences as expertise can aid ML developers in being reflexive and designing technological tools that apply a youth and social justice lens. This chapter will describe opportunities, challenges, and recommendations for future research for engaging marginalized youth and their communities in developing ML-based models to detect child abuse and neglect. Additionally, this chapter will provide recommendations to promote meaningful youth participation in the ML design, development, and deployment process.