The AI Force the United States Needs is a Cyber Force

On September 19, President Trump announced on Truth Social that he was forming an AI Force, “much like I did Space Force,” with an AI czar to follow. On October 4, the czar arrived, and with him a White House task force named the “Super Intelligence Force,” chaired by Director of National Intelligence Jay Clayton and given 120 days to provide a report to the president on AI’s risks and benefits. Whether that task force is the promised AI Force, or a placeholder for one, the White House has not definitively said.

The issue with the Space Force comparison is that AI is not a domain unto itself: It is a technology every service will use, the way every service uses radar. AI does not lend itself well to a structure like the U.S. Space Force, a military service, which has its own personnel system, budget, training pipeline, and a chief who answers for readiness. There is, however, a warfighting domain that is in desperate need of this structure: cyberspace.

Preparing to Fight in Cyberspace

Cyberspace is the first battleground in which AI will change the shape of the fight. In November 2025, Anthropic disclosed that it had detected what it assessed to be a Chinese state-sponsored group using Claude Code to run, by the company’s account, 80–90 percent of an espionage campaign on its own, with human operators stepping in at only a handful of decision points. Last month the company reported a cell in northern Yemen using the same tool in place of software engineers to write guidance code for a guided rocket and a ballistic missile program, and actors in China using it to draft a fire-control specification for an anti-torpedo system. The Yemen cell test-fired the rocket, which failed; within hours, the group was back on Claude asking why. Anyone with an internet connection and an account now has access to that same capability.

The militaries that organized for cyberspace a decade ago are the ones best positioned to use it now. In 2015, the People’s Liberation Army pulled its cyber units into the new Strategic Support Force, combining space, cyber, and information strategic capabilities under a single organization to better integrate these essential elements of modern, digital warfare. In April 2024, Beijing dissolved that force and stood up the Cyberspace Force as an independent arm reporting directly to the Central Military Commission, an organizational move some experts believe demonstrated central party interest in directly managing these capabilities. Israel’s Unit 8200 produces operational outcomes with selectively trained conscripts in their late teens and early twenties, and its alumni have built a security and AI industry whose scale is exemplified by one deal: Google agreed to pay $32 billion for Wiz, a company founded by four Unit 8200 veterans. Where China has built the military organization and Israel has built the industry pipeline, the United States has built neither around the cyber domain.

The Space Force comparison is the most useful parallel the president has provided on the subject, because the history of U.S. readiness for the space domain is similar to where the country finds itself on the cyber domain today. A congressionally chartered commission reported in January 2001 that the Pentagon and the intelligence community were “not yet arranged or focused to meet the national security space needs of the 21st century”; that no service held statutory responsibility to organize, train, and equip for space; and that the United States was “an attractive candidate for a ‘Space Pearl Harbor.’” Its summary of the warning signs was nine words long: “We are on notice, but we have not noticed.” September 11, two wars, and 18 more years passed before a Space Force was created to establish U.S. warfighting dominance for the space domain in December 2019. The clock on the cyber domain has run even longer: The Joint Chiefs listed cyberspace as a domain in the 2004 National Military Strategy, and the Pentagon established Cyber Command in 2009 and elevated it to a unified combatant command in 2018. But in 2026, cyberspace is still the only domain without a service.

Successive commanders and secretaries have tried to normalize cyberspace inside the existing military services. CYBERCOM 2.0, the Pentagon’s current reform initiative, is the latest attempt, but its first concrete product—a new pay system—undercut the effort the week it took effect. The Cyber Mastery Incentive Pay program replaced the services’ separate cyber incentives on October 1, 2026, and was presented as a way to reward operators for advancing their skills. An internal Army comparison reviewed by DefenseScoop showed that nearly every basic and senior skill level across five work roles lost money or stayed at zero, with cuts of $100–1,000 per month, and Bloomberg put the loss at as much as $12,000 a year. The Army has reportedly had to adopt a one-year rate protection for its operators to cushion the blow.

As things stand today, Cyber Command sets the standards and defines the work roles, but as a combatant command rather than its own military branch, it cannot recruit, promote, or manage the career of a single service member. CYBERCOM 2.0 is trying to fix a problem that belongs to four separate services it does not control.

Establishing a Cyber Force

These authors know what the best alternative looks like because they helped write it. The Commission on U.S. Cyber Force Generation—convened by CSIS with the Foundation for Defense of Democracies and on which both authors served—published its recommendations in June. It called for a Cyber Force of roughly 30,000 people that recruits, trains, equips, and retains the force while the military’s combatant commands, to include Cyber Command, leverage and operationalize these forces. The commission report’s estimated $10–11 billion cost for the new service is, in fact, dollars already being appropriated today, but distributed to four services (i.e., the Army, Navy, Air Force, and Marine Corps) for the same mission, rather than new spending. The envisioned Cyber Force adds a Cyber National Guard as a reserve component and gives the service its own intelligence, legal, and education organizations. A future cyber service would necessarily train and educate a workforce fluent in how the president’s priority technologies—currently AI, as well as quantum and other emerging areas—are reshaping the cyber domain. It would also provide an attractive opportunity for top industry experts to contribute for a time to government service.

Like the opposition to the establishment of both the U.S. Air Force and the Space Force, a primary objection to a new Cyber Force is that the current division of labor between four services with different priorities can manage the demands of this domain. This point has been proven false both historically and qualitatively. The creation of the Air Force and the Space Force is viewed objectively as positive developments for the joint force, and regrets are seemingly impossible to find. When looking just at the last 15 years of the military’s cyber experience, there is little reason for optimism: The Pentagon has been unable to resolve the same core issues for more than a decade and a half, and recent initiatives, such as implementing standardized incentive pay, are falling short in real time.

U.S. forces are in contact with the adversary in cyberspace every day, in peacetime as much as in war, and these authors have yet to hear a persuasive case for why it deserves a service less than the other four defined military domains. The president can act now, as he did for Space Force in February 2019, by directing the Pentagon to draft the legislation establishing a cyber service. Congress presented a vehicle in the fiscal year 2027 National Defense Authorization Act, in which Senator Kirsten Gillibrand (D-NY) proposed establishing the force under the Army, one of the two paths the commission laid out. Representative Pat Fallon (R-TX) has said he would like to see this done because a Cyber Force is inevitable.

The president reached for the Space Force comparison because the model is sound. A neglected warfighting domain was given a service, and the service was given authority over its own people and accountability for their readiness. Cyberspace now sits as the only warfighting domain without a dedicated uniformed service, a difficult logic to justify given the transformative impacts of AI on society just in the past 36 months. The cost of waiting is now visible in a pay reform that put retention of a specialized workforce at risk on its first day and in decades-ago planning by an adversary that reorganized for this domain. The task force the president announced will spend 120 days studying AI, but comparing it to Space Force fails to acknowledge the warfighting domain still without a service: cyberspace. AI is and will continue to be an essential capability that each service uses, but its effects today are primarily relevant in cyberspace. If the president still wants a military force designed in the image of Space Force, the best proposal to achieve his stated goals is the Cyber Force these authors helped design the playbook for, one with AI and emerging technologies at its center. That is the force this moment requires.

Joshua Stiefel is vice president for government relations at Second Front and serves as the cochair of the Commission on U.S. Cyber Force Generation, as well as a senior associate (non-resident) with the Strategic Technologies Program at the Center for Strategic and International Studies in Washington, D.C. Rob T. Lee is chief AI officer and chief of research at the SANS Institute and a commissioner on the Commission on U.S. Cyber Force Generation.