The Pentagon should audit its labs and their projects to make sure scientists aren’t developing technology that can be bought off-the-shelf or spun up by startups, according to a recommendation from the Ronald Reagan Institute’s latest National Security Innovation Base report card.
“The review would specifically be focused on, given all the emphasis on commercial first, ensuring that those organizations and that funding is properly aligned and not duplicative of, or in some cases, competitive of what commercial industry is doing,” Eric Snelgrove, a subject matter expert who contributed to the report, told Defense One. “If you look at why the service laboratories exist in the first place, oftentimes it’s a very unique mission…If it’s a 20-year project, private capital is probably not going to fund a private entity to conduct that research.”
The proposed review would focus mostly on the military services’ laboratories, federally funded research and development centers, and university-affiliated research centers, with the Pentagon’s newly reorganized Science, Technology, and Innovation Board leading the charge. There’s also, of course, a place for the Congressional defense committees to oversee the effort.
“How can the labs enable the private sector to move faster, and how can we better leverage all of the private capital that’s being funneled to private companies to deliver capability,” Snelgrove told reporters at a Defense Writers Group event. “And maybe that means, expanding the use of government hypersonic wind tunnels and making those more accessible to industry. Maybe it’s [providing] compute resources for private industry, but making sure that the government laboratories are enabling the private sector, and, again, not competing with them.”
Moreover, the idea is to make the labs, and funds they get, as effective as possible.
“You always need independent experts. You always need tests and evaluation,” Snelgrove told Defense One. “But given all the incentives and the reform that’s been going on to attract more commercial entrants—and the success of a lot of those initiatives—now, I think it’s time to re-look at what the public sector R&D looks like.”
The concept begs at least two more questions: Does the Pentagon even know all of the research projects under its purview and how they’re going? Is the woeful valley of death the result of a fundamental disconnect between deep scientific research and the rapid development in industry?
Welcome
You’ve reached the Defense Business Brief, where we dig into what the Pentagon buys, who they’re buying from, and why. Send along your tips, feedback, and song recommendations to lwilliams@defenseone.com. Check out the Defense Business Brief archive here, and tell your friends to subscribe!
Notes from McAleese. Speaking at the annual defense-programs conference run by McAleese & Associates in Arlington, Va., Emil Michael, the Pentagon’s chief technology officer, said removing Anthropic from the Defense Department’s networks would be minimally disruptive.
- “They all have different strengths,” Michael said of the different generative AI models the Pentagon is using. “The idea was to present all of them to the department. They may all converge in capability if you get this recursive learning sort of concept going in these models. But for now, we need to have more than one option, and ideally all options, and then maybe marry them over time. And I’m pretty confident.
- Update on rollout: “We’ve already deployed OpenAI in the last few weeks, and…we have deployed [Google’s] Gemini. So as these things move up echelon into different classification networks, the warfighter is going to have tons of different options. And what we’re seeing so far is the workflows are very similar. So the disruption is, we think, minimal,” Michael said Tuesday.
- On LUCAS drone production: This is a “relatively new thing for us, so we haven’t gone through the process of what the goal is” with the aim to “mass produce them in this country and have surge capacity so that when we need them, we can create more quickly without having to wait.”
- Bonus round: Budget folks in the Pentagon are largely “pencils down” on the 2027 budget, said Jules Hurst, who is performing the duties of comptroller and chief financial officer. The goal is to have the detail-laden justification books out by April.
Sweet Alabama manufacturing. Raytheon just finished a $115 million expansion to its Redstone Missile Integration Facility in Huntsville, Ala. The move adds 43,000 square feet and increases the facility’s “integration and delivery capacity by over 50 percent,” a company spokesperson told Defense One.
- The new space boasts two new test cells, a bigger factory and dock, and more office space.
- There will be robots! This year, Raytheon plans to add to and upgrade its mobile robot fleet that helps transfer missiles and other items in the factory, called automated guided vehicles. Having more robots means items can be moved faster and, ideally, keep up with higher production needs.
- One to watch: AI-factory company Hadrian is expanding its foothold with a new Navy partnership that includes a 2.2 million square-foot facility to boost shipbuilding capacity in northwest Alabama. I interviewed CEO Chris Power last year, when he previewed expansion plans.
- A related tangent: Anniston Army Depot increased wages for about 1,500 employees. This matters because workforce challenges, namely pay and experience, are key obstacles to production and maintenance of military weapons and equipment.
In other news
- Anthropic gets legal boost from civil liberties groups. The Center for Democracy and Technology and the American Civil Liberties Union filed an amicus brief to support Anthropic in its lawsuit against the Pentagon for labeling it a national security risk. Background: The Defense Department issued a memo earlier this month to remove Anthropic from all systems and networks in the next 180 days.
- Havoc AI is adding aerial and ground autonomy to its portfolio after buying two companies: Mavrik and Teleo. This tracks per my last conversation with CEO Paul Lwin when he sketched out the concept where “one person can control maritime drones, aerial drones, ground drones, and make it all do something sophisticated.”
- Ursa Major flight test of the Draper liquid rocket engine with the Air Force Research Laboratory hit “supersonic speeds” and demonstrated how the Air Force can “leverage our acquisition models to rapidly deliver critical technology advancements,” AFRL Commander and Air Force Technology Executive Officer Brig. Gen. Jason Bartolomei in the news release. Ursa Major CEO Chris Spagnoletti said it took just eight months to go from “contract to flight-ready” propulsion system.
- U.S. drone company Vector signed a memorandum of understanding with Saudi Arabian company SR2 Defense Systems to make, assemble, and upkeep systems in Saudi Arabia. Vector was a competitor in the Pentagon’s Drone Dominance program.
