As well as a few operational successes like Project Maven, which made use of AI for warfighting, the Department of Defense ( DoD ) has spent years diligently studying, researching, and experimenting with it. However, the moment is now for more boots on the ground. The office needs to start digging through the mud to level the execution of extra high-priority use cases in order to promote the adoption of AI.
The establishment of the AI Rapid Capabilities Cell ( A I RCC ) was announced at the end of last year by Radha Plumb, the Pentagon’s chief digital and AI officer, to accede to the department’s adoption of frontier and advanced AI capabilities. The AI RCC wants to enable the department to move more quickly to capitalize on emerging technologies, like conceptual AI, and spend more seriously to scale basic technologies that support AI adoption across the DoD by targeting a specific set of use cases and precise initiatives aimed at putting advanced AI in the hands of warfighters.
The United States has a strategic advantage on the international level for years by deploying sophisticated technologies on the field before its adversaries, according to Plumb. However, it is now being challenged about supremacy and the benefit it provides. She stated that” AI implementation by adversaries like China, Russia, Iran, and North Korea is accelerating and poses a major threat to national security.”
China, for its part, wants to imitate U.S. innovation and import it at prices that American companies can’t match. In a recent blog post, Microsoft Vice Chair and President Brad Smith outlined the higher stakes in this global competition between different regional approaches to AI.” The Chinese correctly recognize that if a country standards on China’s AI platform, it likely will continue to rely on that platform in the future,” said the Chinese.
The DoD must morally incorporate AI into important journey apply cases without delay, according to Derek Strausbaugh, senior director of Microsoft’s National Security Group, in order to stay ahead of competitors and keep its global dominance.
” We are in the AI era.” You can see the investments being made all over the world, both by our friends and our political rivals, by market and by both. Simply put, we need to move more quickly and intently.
Decision Benefits From the Front Lines to the Back Office
The AI RCC has a good program, but the benefits are obviously found in the implementation. With the collaboration between the CDAO and the Defense Innovation System, it is encouraging to see these efforts come together under one overcoat,” said Strausbaugh. He noted that the use cases for AI RCC include possibilities for warfighting and what is known as business management, back office places where lessons from the business use of AI can be quickly and directly applied: logistics, economic management, software development, and others.
He noted that” there’s a great opportunity to move out on many of these very quickly.”
” Planning is one use case for the use of generative AI capabilities,” according to Strausbaugh, which incorporates both large language models ( LLMs) and agentic orchestration of AI workflows. To build foundational models, artificial intelligence ( AI ) systems can quickly process large sets of external data, including global news sources, weather, social media, and scientific papers. integrating domain-specific data from internal communications and military systems to manage and track logistics, readiness and intelligence, etc. can then improve the model to help analysts fully understand the planning process for the military information domain.
No human can reasonably be expected to do that in an expeditious manner if you’re talking about fully developing a complex planning or battle management scenario with all the pertinent information at our disposal. A geopolitical rival will consistently outmaneuver us with more agile decision-making based on a superior information flow built on the principles of speed and accuracy.
A grounded AI can increase human augmentation by creating draft plans for planners to use their creativity and expertise in evaluation and ultimately create a logically sound, doctrinally sound set of plans for commanders in less time than it does joint planning with outdated technology, according to Strausbaugh.
The AI planning system can become more adaptive to change by combining generative AI and agentic orchestration.
and take into account the changing conditions in the battlespace. Agents who can independently evaluate fresh or changing data inputs, publish and collect potential orders, and use distributed battle management where every human planner and commander operate from the most recent plans and understanding of the battlespace are.
Strausbaugh added that the AI system should complement, not replace, human decision-making. The planner “makes the decisions,” he said, and in doing so, they can also provide feedback on the AI inputs that can be used to further enhance and fine tune the model.
The AI RCC is also moving forward with other generalizable enterprise or business system functions, such as software development or financial management.
Modern AI Infrastructure
Some might think that this deployment shift is overdue, but the long runway, all those studies and experiments, have given DoD customers a thorough understanding of the kind of infrastructure they need for AI operations, and it’s accelerating a new, highly developed wave of cloud migration, according to Strausbaugh.
” Cloud 1.0 was essentially]DoD customers ] asking how do I cut costs,” he said. How can I increase my operational efficiency? How can I use platform as a service offerings to avoid having to buy and maintain hardware?” he stated.
However, he said, the drive to cloud in the DoD has changed from a push move ( customers seeking to move to maximize savings ) to a pull one ( customers seeking to move to maximize value ) more recently.
DoD customers in this second wave see cloud migration as a way to realize the extraordinary potential value and utility of their data to improve decision-making and improve operational effectiveness. They are moving portfolios of data and applications into the cloud in cloud 2.0 because they can benefit from having a huge advantage by using specific data sets relevant to particular DoD missions ( for example, logistics ) to enhance foundational LLMs and create generative AI systems that can really increase productivity and speed up decision-making.
Although it’s a question of whether AI programs need a lot of cloud resources, Strausbaugh said. Customers come with sophisticated inquiries that are largely based on experimentation and some experience. How can we best organize our data to ensure that it retains its value, particularly in the current context? How can we use that data to make it AI-ready by performing the best operations? Do you ultimately have the hardware we need to create and retrain models and run effective AI applications on? he stated.
According to Strausbaugh, the Department of Defense is at a critical juncture with AI:” The general population’s familiarity with ChatGPT and other similar implementations has created a curiosity and willingness in the department to understand and engage in a very healthy dialogue about how to best capitalize upon these emerging technologies. In order to reduce the cognitive load on our warfighters, industry and government should both automate the mundane and maintain the human-in-the-loop and on-the-loop, where generative and agentic AI systems can assist in improving the speed, quality, and accuracy of decisions being made by our nation’s most valuable asset, our people. There are an incredible opportunity and responsibility ahead of us.