The U.S. Air Force's strategic advantage, particularly against China, is threatened by advancements in machine learning, necessitating new approaches to mission planning. This report explores a prototype AI system designed to accelerate and enhance the development of operational concepts for air domain missions. By integrating open-source deep learning frameworks, contemporary algorithms, and a U.S. Department of Defense-standard combat simulation tool, the system aims to leverage AI's capacity for large-scale learning, generalization, and iterative improvement. The authors investigate the potential of AI agents to orchestrate collaborative behavior in simplified suppression of enemy air defenses missions, highlighting both the promise and the challenges of reinforcement learning in this complex domain.

This report presents a preliminary exploration into the application of AI and ML for enhancing air dominance through improved mission planning. The authors detail a prototype system that integrates advanced computational tools to address the growing threat posed by competitors leveraging similar technologies. Initial findings suggest that reinforcement learning holds significant promise for tackling complex, collaborative air mission planning scenarios, while also acknowledging substantial challenges that need further investigation. The work serves as an important step in understanding how AI can be leveraged to maintain a strategic advantage in air warfare.