The Chief Digital and Artificial Intelligence Office (CDAO) said its goal was to accelerate the Department of Defense’s (DoD) adoption of agentic AI capabilities to “address national security challenges.”
CDAO head Doug Matty said the Defense Department wished to tap into the best technologies developed by U.S. AI companies to support its troops and maintain a strategic advantage.
The U.S. military is no slouch when it comes to technological innovation. For example, the DoD’s R&D arm — Defense Advanced Research Projects Agency or DARPA — created Arpanet in 1969, a communications network that linked computers far apart. Arpanet later became the internet.
DARPA has contributed to the field of AI over the decades, backing everything from expert systems to autonomous vehicles. In 2023, it developed an AI system that autonomously piloted an F-16 and engaged in dogfighting scenarios with a human-piloted F-16. A year earlier, it debuted a Black Hawk helicopter piloted only by an AI system.
Commercially, enterprises are developing AI-powered products to sell to the military, including autonomous or semi-autonomous drones, surveillance and reconnaissance systems, targeting systems, signal intelligence, flight control and decision systems and the like.
AI is already being used across the military. Last month, the U.S. Air Force completed “Experiment 3,” which tested the pairing of human and AI systems to speed up responses to threats known as the “kill chain.”
One of the tools tested was the Maven Smart System, which used AI to give real-time suggestions to military teams about possible targets and actions. Humans still made the final call. The goal was to see if AI could enhance human decision-making — not to replace the people.
In March, the DoD’s Defense Innovation Unit kicked off “Thunderforge,” an AI-driven military planning effort co-developed with Google and Microsoft. The AI system integrates intelligence streams and battlefield sensor data to generate operational recommendations with humans in control.
Read more: PYMNTS Intelligence: Finding the Way to Better Government Disbursements
Risk Worth Embracing?
The military’s embrace of AI signals a realization that 21st century warfare is changing and the U.S. military must modernize to keep up.
“The U.S. ability to deter war in regions of critical interest is fading,” wrote Carol Kuntz, adjunct fellow at the Center for Strategic and International Studies, in a June 2025 report titled “Artificial Intelligence and War.”
“Adversaries — particularly China — have improved their own capabilities and could now threaten classic U.S. power projection.
“AI-enabled military capabilities, particularly sensor and firing networks, combined with other force and program improvements, would be among the most promising strategies for shoring up deterrence and defense capabilities in the near- and mid-term,” Kuntz wrote.
But alongside the realization about the necessity of AI for warfare is a concern that these systems could make mistakes in judgment.
“If AI is to be used in war, decisions in the command center or the Situation Room would need to rely on predictions about the effects of the use of algorithms, particularly in sensitive applications,” Kuntz said. “Absent such predictive tools, AI-enabled military capabilities could not be responsibly authorized in uses such as a sensor and firing network.”
According to a 2024 Stanford and Georgia Tech paper that tested five language models in military and diplomatic decision-making, researchers found that all models took actions that escalated the fight.
“We observe that models tend to develop arms-race dynamics, leading to greater conflict, and in rare cases, even to the deployment of nuclear weapons,” the authors wrote.
The models justified their actions by saying they wanted to deter enemy actions by taking the first strike, according to the paper.
Nonetheless, AI model developers are forging ahead to sell their technology to the military. They have created dedicated business units to cater to the needs of the military and U.S. government. There’s Microsoft’s Azure for U.S. Government, Google for Government, Amazon’s AWS GovCloud, OpenAI for Government, xAI’s Grok for Government, Anthropic’s Claude Gov, among others.
In the end, despite AI’s inherent risks when deployed in warfare, the U.S. may not have a choice but to embrace the technology, especially since adversaries are not standing still.
“The United States is confronting the rise of a peer competitor, as well as a host of other military dangers and problems,” Kuntz said. “China fields precision guided munitions, hypersonic missiles and fighter aircraft increasingly able to pierce U.S. air superiority, capabilities that pose risks to classic U.S. power projection forces.”
“There is a reasonable basis to believe that AI-enabled military capabilities could help rectify many deficiencies in U.S. combat power,” she added.
Read more:
Visa: Digital-First Governments Use Payments Data to Improve Policy and Financial Access
Point & Pay Integrates Trustly’s Pay-by-Bank Tools for Government Payments
New Report: Instant Digital Disbursements Cut Through Government Red Tape