Pentagon deploys AI to predict China’s multi-vector cyber attacks

The Department of Defense plans to deploy artificial intelligence to predict and simulate adversary attacks as part of a major network consolidation effort under U.S. Indo-Pacific Command. The initiative, called Mission Network-as-a-Service, aims to unify 17 disparate networks into a single secure system while using AI to enhance cybersecurity defenses against sophisticated threats from adversaries like China.

What you should know: Katie Arrington, performing the duties of DOD CIO, unveiled the Mission Network-as-a-Service program that will eventually extend to all combatant commands.

• The initiative consolidates 17 separate networks used by Indo-Pacom to communicate with allies into one unified system.
• It supports the Pentagon’s broader Combined Joint All-Domain Command and Control (CJADC2) vision to connect military sensors and weapons under a single network.
• The unified networks will be built with zero-trust security principles and feature reduced attack surfaces.

How AI enhances defense: Artificial intelligence will serve multiple critical functions in protecting the consolidated network infrastructure.

• AI will conduct predictive analysis to anticipate and thwart adversary attacks before they occur.
• The technology will help distinguish between genuine outages and orchestrated cyber maneuvers from hostile actors.
• Machine learning models will be trained to think strategically about making adversaries spend more resources on their offensive capabilities.

The threat landscape: Arrington emphasized that potential adversaries like China employ sophisticated, multi-vector attack strategies that are difficult to detect.

• “They do not intend to go after one attack path. It’s multiple at once, and it’s going to be nondescript. You’re not going to be aware it’s happening,” she explained.
• Adversaries condition defenders to accept small disruptions in power, network connectivity, or financial institutions as normal occurrences.
• This conditioning makes it challenging to identify when incidents are part of coordinated attacks versus routine technical issues.

The economic warfare strategy: The ultimate goal involves using AI to create a cost-asymmetric advantage in cyber warfare.

• “If I’m building a capability to defend, the offensive capability needs to be less money, more accurate,” Arrington said.
• The strategy focuses on forcing adversaries to spend disproportionately more resources protecting themselves.
• AI models will be trained with the principle: “Whatever you do, make my adversary spend more, waste more time, spend more time trying to afford it so that I can make a leapfrog ahead of them.”

What they’re saying: Arrington framed the initiative as fundamentally an economic competition enabled by artificial intelligence.

• “That’s the game that we’re actually in right now with AI: How we build a capability that can beat the whack out of them and make them spend more to protect themselves against it. That’s where you win the war.”