How Generative AI Improves Military Decision-Making
The U.S. military’s use of generative artificial intelligence (GenAI) gives decision-makers the ability to not only train against predictive enemy tactics but also prepare for a multitude of hiccups that could occur during operations.
Leigh Madden, vice president of the National Security Group at Microsoft, delved into the military benefits of adopting GenAI tools during a SIGNAL Media Executive Video Series episode. In his role at Microsoft, Madden is responsible for all aspects of Microsoft operations for customers in the U.S. federal market.
GenAI can process intelligence, signals and reconnaissance data in real time, enhancing decision-makers’ situational awareness, Madden said. It can also be used to generate operational scenarios and recommend effective strategies.
AI-driven war games offer commanders and warfighters immersive training experiences. For example, Madden said GenAI training scenarios incorporate different terrains, weather conditions and enemy tactics and behavior.
“This is really about pairing between AI and an operator. It's not about replacing the human,” Madden emphasized. “AI is used more as a decision aide by improving both speed and precision and applying responsible AI principles to guide the use of AI within implementation-specific guardrails.”
According to Madden, AI can also be used to analyze performance data from the training exercises and provide feedback for users so they can learn and strengthen their decision-making abilities.
To ensure the security of AI systems against cyber attacks, the AI models get exposed to adversarial examples during the training process, Madden said. Additionally, robust input validation mechanisms within AI can monitor unusual input patterns and recognize and reject malicious inputs to deflect adversarial cyber attacks.
Madden also said the data used to train and operate AI systems must be encrypted, and regular audits must be conducted to ensure the operational success of AI systems.
Microsoft’s tools used within its own cyber defense operations, including GenAI models, are available to federal customers, Madden said.
“Generative AI is used in developing intelligent agents that continuously monitor network traffic for anomalies and indicators of compromise, as well as understanding adversary tactics,” Madden explained. “The agents can then automatically respond to the detected threats, reducing response times and mitigating potential damage.”
Madden also said GenAI systems can predict maintenance schedules for military equipment by analyzing sensor readings, maintenance logs, historical performance data and environmental conditions. AI can predict equipment failures before they occur and simulate different failure scenarios.
“This proactive approach ensures that maintenance is performed only when necessary, reducing unnecessary downtime and optimizing resource allocation,” Madden said. “Related necessary parts and experienced maintainers can be prepositioned so that supply chain constraints or lack of expertise does not take a platform out of service.”
