The Pentagon wants to infuse the U.S. Cyber Command with a massive increase in funding to expand its AI use. The 2027 CYBERCOM budgetary plan would jump 27 times over its 2026 allocation, increasing from $5 million to $138 million.
As first reported by Breaking Defense, the funding increase would seem to herald a marked increase in cyber operations, especially missions that are offensive in nature. AI would be a key tool in determining targets, automating parts of mission planning and analysis, and generally speeding up workflows. From an intelligence and surveillance standpoint, AI would help “cyber warriors” process large volumes of data more quickly, identify suspicious activity and track adversaries.
Funding also would be provided for foundational AI infrastructure like cloud systems, generative AI models, LLM access, agentic AI capabilities and retrieval-augmented generation (RAG) frameworks that encourage an AI to access “open book” sources both internal and external to the Pentagon and to provide source links in order to reduce the chances of an AI hallucination.
CYBERCOM’s profile appears to have increased following U.S. removal of Venezuela President Maduro on January 3, 2026. As reported by Cyberscoop, Paul Lyons, principal deputy assistant secretary for cyber policy in the Pentagon, recently waxed enthusiastic about CYBERCOM’s shift from defense to offense.
“We saw it in spades in Venezuela,” said Lyons, “where you can layer cyber to create conditions that are favorable to the warfighter, that lower the risk to mission, lower risk to force that where paired with both non-kinetic and kinetic effect, can increase lethality.”
Frontier models like Anthropic’s Mythos AI, with which the Pentagon has a testy relationship, represent a “watershed moment” in potential offensive missions, said Lyons. “America’s posture in cyber defense has been largely a defensive posture. That’s a losing strategy for America. America has to dominate the full spectrum of cyber operations.”
The Pentagon now fields a “non-kinetic effects cell” that is pushing cyber operations to the forefront of military operations. The goal is to influence or disrupt an enemy’s systems without using physical force or direct destruction. This is a step up from previous signals interception tasks that would indicate troop movements or radar activation, for example. That combat marriage between non-kinetic and kinetic was well demonstrated in Venezuela where the country’s air defense systems were effectively dismantled before it could fire a shot.
Similar cyber operations also were among the “first movers” in operations against Iran, according to the Pentagon. The Pentagon sees its ties to industry as a vital part of this initiative as that’s where these new AI tools originate. That may indirectly include Ukraine as Pentagon-favorite Palantir has sealed a deal that gives it access to Ukraine’s trove of Brave1 battlefield data. Palantir’s Maven AI is a critical enabler for the Pentagon.
While the Trump administration emphasizes aggressive cyber operations, CYBERCOM also must focus on its mission to defend the “digital homeland.” The budget includes funding for AI analysis of malware, quicker threat detection times and automated reports of cyber threats as they become more machine-driven. There are evident vulnerabilities. Chinese hackers, for instance, reportedly still have a presence in American networks.
The defensive skills of CYBERCOM also came into high relief in the wake of the March 11 cyberattack against Stryker, a U.S. medical device manufacturer that operates globally. The destructive “wiper” attack with no ransom demand against tens of thousands of devices was allegedly carried out by an Iranian-linked group called Handala Hack (aka Void Manticore) and is being called the most significant wartime cyberattack against the United States in history.
