This book chapter, a conference paper presented at the Association for the Advancement of Artificial Intelligence 2017 Fall Symposium, describes an initiative to create an artificial intelligence (AI) decision aid to improve current coordinated housing entry systems for homeless youth. Using data from the U.S. Department of Housing and Urban Development (HUD) Homelessness Management Information System (HMIS), the researchers first determined whether the current rubric used for prioritizing youth for housing assistance is an effective predictor of youths’ probabilities of success. Next, they worked to build better AI decision aids and predictive models using components of this rubric. The study analyzed the most frequently used risk/vulnerability assessment tool called the Next Step Tool (NST) for Homeless Youth. Their analysis indicated that assigning youth based on NST scores is an effective intervention for assisting high-risk youth. Moreover, the researchers believe additional predictive analytics and AI decision aids can augment assignment decisions based on NST scores. This study found much potential for effective human-machine collaboration in the context of housing allocation. The researchers plan to work with HUD and local communities to develop such systems in the future.