AUSTIN, Texas—Despite protests and a sponsorship ban, defense tech still has a foothold in the Lone Star State. And the evidence was clear this week at the edges of the colorful and always festive South by Southwest conference.
Much of the activity was centered at the Capital Factory, home to several defense tech organizations including the Pentagon’s Defense Innovation Unit, which hosted a two-day conference that began March 7.
After that event, defense officials were somewhat scarce due to travel restrictions, despite having lined up speaking engagements at subsequent conferences. Many of those who did make appearances were local, such as Lt. Gen. Eugene Brown, the second-in-charge of Army Futures Command; Brig. Gen. Jason Bartolomei, the leader of Air Force Research Laboratory and AFWERX, which has an office in the Capital Factory; and Devin Beckwith, Special Operations Command’s technology liaison. All three spoke at the two-day Tectonic Defense Summit that started on March 10 at the South Congress Hotel.
Across days of discussions, a theme was clear: to gain military advantage, the Pentagon must take more risks—particularly when it comes to drones and AI.
“I think there will be a time where we get autonomy that we can trust to do almost exclusively everything that we want it to do. But until that time, we don’t need to wait for a perfect product. Let’s take a [minimally] viable product. Let’s get it out in the field. Let’s start to use it so that people build trust,” AFC’s Brown said Monday.
Software buzz
There was also plenty of chatter about Defense Secretary Pete Hegseth’s directive for the Pentagon to buy software faster—whether it’s for weapons or desktops.
“We have to take the risk,” Lt. Col. Nick Estep, who leads the emerging technology portfolio for the Defense Innovation Unit, said at Maxar’s Orbital Edge event Sunday. “The alternative is so much worse.”
DIU was one of the first defense organizations to use the software acquisition pathway, which was released in 2020, as a standard way of doing business, Estep said. But having the directive come from the defense secretary sends a “pretty strong and potent message” that “we are going to look at fielding software in a much smarter fashion.”
Estep stressed that the change will mean accepting imperfection and failure.
“We have to field it. We’ll have to show joint experimentation and learn,” he said. “I think even the venture capital arms understand that there is risk. Not everything is gonna be extremely successful, not everything is gonna materialize in some meaningful contract at the end. So that’s just a part of the culture and part of the acquisition process.”
But while the financial firms are aware of the risks of investing in defense technology, the goal is still to turn a profit.
“I think the problem is: how far can our companies get funding their own stuff and then selling it at high margins before the customer says, ‘Well, we want something different and we want to pay you to develop it and do what our way,’ and then your margins plummet, and you’re trading off, I gotta keep growing the top line, maybe I’ll dilute myself,” said James Cross, the managing director for Franklin Venture Partners and private investing co-lead for Franklin Equity Group, during a panel on investing at the Tectonic Defense Summit on Tuesday.
“We saw Hegseth’s memo about software the other day,” he said. “We’re moving to a software-driven autonomous, attritable battlefield. Well, if you want daily software upgrades, you gotta pay software margins. So, there is hope on profitability, I think.”
All aboard with AI
The Pentagon has repeatedly signaled that it wants to use cutting-edge software and embrace AI and autonomy for operations and in war. But despite some forward movement with programs like Replicator, there are deficits where there could be easy wins.
When talking about incorporating AI responsibly, Estep said the government has done a terrible job broadly implementing IT infrastructure and “just simple business automation tools …to just make our lives better,” he said.
And while it’s important to mitigate AI risks and keep the human in the loop, particularly for mission critical functions, integrating the tech broadly is essential, he said.
“For us to gain an asymmetric advantage, we have to be able to leverage this, you know, the most advanced capability out there, its horizontal integrator. You can’t avoid it. It’s what’s going to allow us to be successful going forward. So we might as well figure out how to adopt it, and integrate it across the entire Defense Department.”
One battlefield use case is having drones ferry supplies instead of crewed ships, boats, or helicopters, which are often targets for adversaries and put troops in harm’s way.
“You don’t want to put a ship in the middle of the Pacific … when you can have autonomous assets sit out there for months at a time and you can have humans monitoring from thousands of miles away, safe,” Paul Lwin, co-founder and CEO of HavocAI, said at the Tectonic event on Monday.
Those autonomous systems could also keep the Navy’s expensive and delayed-production-prone warships from being damaged or destroyed in battle, Lwin said.
“Yes, we should keep building capital warships, and we should also look at the future. But the reality is: 10 of our vessels can take out a destroyer—and our vessels cost $100,000 per vessel,” Lwin said. “You can do the math, right? At the end of the day, warfare is economic. If you can take out a $40 billion warship with $1 million in robotic boats, we’re gonna win. If we go $40 billion to $40 billion and fight that battle, we’re probably not going to win against a country that has an industrial capacity” as high as an adversary like China’s.
But removing bureaucratic barriers and producing those systems at scale in preparation for a potential conflict will be one of the Pentagon’s biggest challenges.
“Our biggest opportunity to make a difference between now and 2027 is more autonomous robotic weapon systems and sensor platforms and decoys,” Nelson Mills, founder and CEO of Vatn Systems, which makes subsurface drones, said Monday during Tectonic’s “Autonomy at Sea” panel.
When asked if the Pentagon’s appetite for acquisition reform could yield real change, Mills was cautiously optimistic.
“I mean, it’s definitely possible. Can it be done? Yes. Will it be done?” Mills said. “I think all the traction and movement is, in general, positive and things are improving. You look at the FORGED Act, you look at what the new administration is trying to do on the acquisition side—all the right pieces and building blocks are there. It’s just about executing and making sure we actually fund these programs. And just make decisions on acquiring systems or a family of systems, and move forward on deploying them.”
Read the full article here
