Military use of drones for air, land, and sea is booming—and so is the need to power them. So the Pentagon is working on a strategy for how it sources and buys batteries—including the critical minerals they require—which officials expect to release in 2026.

This strategy will be an update to a lithium ion battery strategy published in 2023 “that laid out the groundwork for how we were going to address battery challenges in the department,” including working with other government agencies, allies and partners, Eric Shields, the senior battery advisor in the Pentagon’s industrial base policy office, told reporters Friday.

Congress mandated a department-wide battery strategy in the 2025 annual defense policy bill. The goal is to have the new strategy signed by March 2026. 

“What we know is, especially from the battlefield in Ukraine, that batteries are really important. They're important for enabling capabilities like drones, communications, and many other things that we need to fight and win. It's important for the department to have secure supply chains for these and…really important that we have standards, because that's one way that we're going to tackle some of these challenges,” Shields said. “And you can see clearly in executive orders coming down on drone dominance and security, you can see guidance coming down on the importance of critical minerals to the administration that these are priorities and are going to need resources to make progress.”

The Pentagon’s reliance on critical minerals, including those required for batteries, has long been a concern, especially because much of the supply chain resides in China.

Nearly 80 percent of the Defense Department’s weapons systems rely on critical minerals, according to a recent Govini report. And that demand is only increasing as the White House and Pentagon push new technologies, such as drones, across all warfighting domains. 

“We need these batteries yesterday. I need way more energy and power available in all of my platforms than what I've been fielding to date. And and the more I can get out there, the better off we're going to be,” said Daphne Fuentevilla, the Navy’s deputy director for operational energy. 

Friday, the Army, Navy, and Marine Corps hosted their first industry day for battery technology, with a focus on standardizing and improving supply chains and stockpiles. 

“This was really the first time we sat down and collectively hosted an event with a specific purpose of bringing industry into the discussion as an equal partner,” said Marnie Bailey, the senior scientific technical manager for power and energy at the Army’s Command, Control, Communications, Computers, Cyber, Intelligence, Surveillance and Reconnaissance Center. 

“Technology is moving fast. The Army is moving fast, the Navy is moving fast, DOD is moving fast. And so this industry day was the beginning of this conversation, this open dialogue with industry, so we can move fast together.” 

Standardizing batteries from all angles, including cell material and modular architecture, is a primary focus, since different platforms and devices have different batteries. 

“One of the things that that we're looking looking at standardization for is: if we have to engage in conflict today—and we're not going to have a lot of time to do a lot of the assessments and and review every single individual piece of equipment for the next two years before we field something—we're going to be taking some risk. And the more we can provide this upfront guidance on standardization, the better off we're going to be when we do have to actually be in a conflict,” Fuentevilla said. “Areas where we're looking at future standardization: the small UAS space, our undersea architectures…large format batteries embedded into our ships, is also going to be a challenge.”

Making sure there’s a robust, secure, and domestic supply chain that can keep up with demand is also another priority. 

“Some of this technology evolution within batteries has really enabled us to get after more capable defense weapon systems and platforms. And we have to be able to keep up with that in order to address our mission,” Fuentevilla said. “Specifically, we need batteries in our destroyers and frigates on the surface side. I need more capable sensors and weapons on our surface platforms. I also need small unmanned aerial systems, and these one-way attack drones. Because for these systems in particular, the capability that I can deliver is directly proportional to the amount of energy and battery energy that I can put on those systems.” 

The Navy is also investing heavily in uncrewed surface and undersea vehicles that require battery power. And the Army and Marine Corps rely on battery-powered equipment for mobile equipment, aerial and ground vehicles—a challenge the Defense Innovation Unit has been working on.

One key takeaway from the inaugural event was that companies want to know when and what the Pentagon wanted to buy. Shields said the money is there, but the requirements are still en route. 

“With the significant resourcing the department just received through the one big, beautiful bill, a lot of those decisions are still being made… stand by is kind of the message, because the requirements are coming, but we don't have them just yet.” 

On Thursday, the Energy Department proposed $1 billion in an effort to shore up the critical minerals supply chain. Earlier this year, a White House issued an executive order calling for an increase in critical minerals production, and the recently passed budget reconciliation carves out $1 billion in appropriations for Defense Production Act, which has previously been used to produce critical minerals. 

The bill also includes $2 billion to improve critical minerals stockpiles and supply chains “through the National Defense Stockpile Transaction Fund” run by the Defense Logistics Agency, plus $5 billion for “investments in critical minerals supply chains” for fiscal year 2025. 

]]>

AI is expected to play a central role in the Golden Dome air-defense system, from helping to integrate today’s sensors and interceptors to accelerating the detection and tracking of threats, according to slides shown by Defense Department officials last week at an industry day in Huntsville, Alabama. The slides also provided new detail on the Pentagon’s ambitions for missile-killing satellites and other matters.

More than 3,000 people from the space and missile-defense sectors attended the daylong event, which was held amid—but was not formally affiliated with—the industry-group 2025 Space and Missile Defense Symposium. Defense Secretary Pete Hegseth forbade Defense Department officials to discuss Golden Dome at the SMDS, and reporters were barred from the industry day, whose discussions were unclassified. 

Asked for details on the proceedings, a Missile Defense Agency spokesperson referred Defense One to the Office of the Secretary of Defense, which sent an email saying, in part, “The Golden Dome for America office is examining current and future solutions across the services and interagency to identify the most effective ways to modernize and quickly field the capabilities our nation needs to protect our homeland.”

Defense One obtained a copy of the slides, whose authenticity was confirmed by several attendees, including a government official.

An AI-powered Golden Dome

New automation and AI tools—including an “AI-Enabled Fire Control Concept”—were a central feature of the industry-day presentations. 

AI is expected to help network a wider variety of radars and missile batteries, and may enable the tracking of far more missiles than is possible today. 

“The reason you need AI help is because, instead of a handful of missiles, or a dozen or so from North Korea or Iran, now we’re talking about what could be dozens and dozens or hundreds from Russia or China. There’s a quantity challenge and then there’s a time challenge,” one attendee said. “You want to be able to hit these as quickly as you can, and AI can sort through that much, much faster than a human can.”

Officials didn’t provide details on what an AI-enabled fire control concept would mean in practice, but some aspects of missile defense, such as target warning, already have AI elements.

Speaking at the Space and Missile Defense Symposium—but not on the restricted industry panel—Dan Wald, director of artificial intelligence at Booz Allen Hamilton, said AI could serve a variety of tasks related to space-based tracking and interception of missiles without necessarily removing humans from oversight of firing on incoming threats. He outlined a vision for AI-enabled fire control where humans play more of a supervisory role than an active one—but still act faster. 

Wald described it as “moving the human from 'in the loop' to 'on the loop' for fire control…This smart fire control can basically give recommendations for intercept or for applications of resources. This is where we can reduce the amount of people that are necessary to close that loop from a dozen to, you know, two men in a can.”

Maj. Dwight Hicks, space strategy officer at Space and Missile Defense Command, said AI could also help with rearming and maintenance, allowing for streamlined logistics. 

“You shouldn’t have to wait for a platoon leader or platoon sergeant to say, ‘Hey, I need more ammunition, or I need more big bullets or beans.’ It should be automated. So if you know a big launcher is shooting missiles, there should be an automated count as that is going, and it triggers for it to go to the rear and the rear to start moving forward.”

Some said AI could also help the program reach its ambitious 2028 goal by speeding up testing. One slide describes a greatly accelerated testing “cadence,” as well as frequent and ongoing tests related to software, ground sensors, and more. And the Pentagon is looking for companies that can conduct more of their own testing, and show that it is rigorous, one attendee said.

“What they’re trying to do is figure out ways to accelerate testing. For example, can you apply AI to try to accelerate some of the data review and analysis? That would be a huge advantage. And also, can you do more piecemeal testing, building up to integrated testing?” they said.

Missile-killing satellites

Even space-based interceptor tests don’t have to be as expensive as they might seem, one attendee said. “If you want to test a space-based interceptor, you don’t have to necessarily launch it into orbit and then test it there. You could do suborbital testing of your kill vehicle with much cheaper launch costs and a much faster schedule.”

The Golden Dome program, at least at this phase, isn’t seeking a single type of space-based interceptor from one provider. One slide mentions the Brilliant Pebbles program, canceled in 1994, which envisioned a constellation of missile-shooting satellites from Lockheed Martin and Northrop Grumman: “Advances in technology, manufacturing, and cost curves since the Brilliant Pebbles program make space-based interceptors feasible — but not simple. The U.S. has never built a re-entry vehicle that can close an intercept.”

The planned Golden Dome interceptors will have to do more than what planners originally envisioned for Brilliant Pebbles. Brilliant Pebbles was conceived in the 1980s as a means to destroy missiles in the earliest (and most-targetable) part of their flight, when they are just lifting off into space.  It was later expanded to include the mid-course phase in the early 1990s. One slide on Golden Dome says that program officials want the new interceptors to be able to down missiles at every stage of flight, from lift off through the mid-course and glide phase, a feature of modern, highly maneuverable hypersonics that was not really a consideration decades ago. 

Attendees said that suggests the program will be open to buying several types of interceptors to provide redundancy. 

Closer than you think

The prospect of space-based interceptors, a concept that has existed for decades but never been effectively deployed, has garnered much attention. But the main thrust of the Golden Dome architecture, as laid out in the industry day presentations, will be tying together a wide variety of radars, sensors, and missiles already in use or in development, such as Northrop Grumman’s Ground-Based Midcourse Defense system, Lockheed Martin’s Patriot PAC-3 air and missile defense system, the in-development Next Generation Interceptor, the Integrated Battle Command System, and other  systems from Boeing and Raytheon.

Pentagon plans call for bringing these together, developing new “common launch” missile batteries, and deploying 11 short-range missile batteries throughout the United States, according to multiple slides.

Golden Dome also will work much faster than today’s missile-defense regime, the slides say. Today’s architecture is challenging because of the lags in communication “across the kill chain,” whereas officials want the new system to include “Next-Gen” attributes such as  “seamless integration with joint assets—any sensor, any shooter” and “integrated left-right of launch,” meaning intelligence collection and sharing long before an adversary missile launch has occurred.

That integration is the biggest challenge, attendees said: coordinating the data from sensors in various locations, on land and in space, and ensuring compatibility across a wide variety of launch systems from multiple vendors. “How do you command and control all of that? That’s the hard part, especially when you’re talking about thousands of space-based interceptors as well as a growing number of ground-based radars and missile systems,” one person said.

It will require more than just existing missile defense pieces. 

Tom Karako, director of the Missile Defense Project at CSIS, declined to speak about the industry day specifically, but pointed to a previous CSIS publication outlining the need to incorporate radar, sensors, and satellites from outside the Defense Department—such as NOAA assets—to increase interception chances, but also complexity.

U.S. defense architecture today, while highly complex, still requires significant manual coordination among the services under extreme time pressure, all taking quick action to train radars or other sensors on new threats, one attendee explained. That’s one reason AI and automation play such a big role in the Pentagon’s Golden Dome plans.

Teams forming

SpaceX, which today is the cheapest launch provider and has launched its own satellite constellation, is considered one of the most competitive companies for Golden Dome contracts. However, it goes unmentioned in the slides and was largely missing from the day's conversations. 

Why is that? One attendee said company officials “haven't shown a real interest” in Golden Dome, aside from the need for lots of launches, for which they are the favorite. “I would be very surprised that they actually try to build a space-based interceptor. I think what they would probably do is try to be on everybody's team.” 

Speaking outside the industry day, a Northrop Grumman official said they are exploring new ways to work with competitors to solve the data and integration challenges and “collaborate across those company boundaries quickly. That’s what’s going to enable Golden Dome.”

Amanda Pound, advanced programs development director at Lockheed Martin, also speaking outside the event, sounded a similar note. “SpaceX has very capable vehicles. But there are also many other launch providers. It’s gotten cheaper, more sustainable over time.”

The Golden Dome program might reshape the nascent space business sector, even diluting SpaceX’s current leadership position. 

“The overall effect of this is to create a bunch of new potential SpaceXs, because now you have the economics of scale behind you to really launch the space economy.”

Why so silent?

The curtain of secrecy around unclassified discussions renewed questions about the controversial program. Experts have expressed doubt about the administration's claims about timeline, projected cost, practicality, and effect on deterrence. 

“‘Golden Dome’ probably sounded good to the president, and now no one is going to talk him out of it—especially given that the administration is willing to throw mountains of money at such a program, just as Reagan did,” Tom Nichols wrote in The Atlantic. “Hegseth can order his people not to talk about it at public gatherings, but at some point, the administration should answer the two most important questions about an expensive system that could destabilize nuclear deterrence: What is Golden Dome supposed to do, and does it have any chance of working?”

]]>