 Antimicrobial resistance is a growing global threat to public health. Computational tools have improved our ability to predict infectious diseases since the early 2000s, but there has been little progress in developing real-time forecasting models for antimicrobial resistant organisms, AMROS. This paper examines the potential benefits of AMRO forecasting at multiple levels, identifies key challenges, and suggests areas for future research. It also highlights the difficulties associated with accessing quality data, calibrating models, and implementing and evaluating forecasts. Finally, it emphasizes the importance of leveraging existing data and resources to begin addressing these issues now. This article was authored by Senpei, Seth Bloomberg, Jamie Cascanti Vega, and others.