America’s long-range kamikaze drone, a reverse-engineered clone of the Iranian-designed Shahed-136, was used in combat for the first time during Operation Epic Fury. Before the Low-Cost Uncrewed Combat Attack System (LUCAS) was known to exist,The War Zone made a detailed case for the Pentagon to mass produced this exact aircraft. Now the U.S. military says LUCAS is a huge success – with CENTCOM’s top officer calling it “indispensable,” – and wants many more of them, we wanted to know more about how these weapons were first developed. One former Pentagon official who pushed for their creation, a drive that began under the Biden administration and came to fruition under Trump, gave us unique insights into LUCAS and its genesis.
Michael C. Horowitz served as Deputy Assistant Secretary of Defense (DASD) for Force Development and Emerging Capabilities and Director of the Emerging Capabilities Policy Office in the Pentagon between 2022 and 2024 under the Biden administration. Now a senior fellow for technology and innovation at the Council on Foreign Relations and director of Perry World House and Richard Perry professor at the University of Pennsylvania, Horowitz spoke to us Wednesday. He provided unique insights into why it took so long for the U.S. to fully embrace the use of a weapon that has proved devastating in Ukraine and now across the Middle East, and how it could be used in the future against China.
Some of the questions and answers have been lightly edited for clarity.
Q: When did the U.S. first obtain a Shahed from Ukraine and what were your thoughts?
A: It was early 2024 essentially. One of the things my office did was look at what were the emerging capabilities that may be in the services and where there might be a lot of potential, but for whatever reason, weren’t getting the support that they needed. And in May 2024, the question of a potential for this came across my desk.
Q: How did Russia’s experience in Ukraine inform your decision about pursuing such a weapon at the time?
A: Given that we were in the era of precision mass and that the U.S. arsenal has been entirely made up of these exquisite, expensive, difficult to produce capabilities, there was interest in finding lower-cost alternatives – more attritable, more autonomous kinds of systems. And at that point, the world was now familiar with the capabilities of the Shaheds from the thousands that Russia had been launching against Ukraine. And Russia was starting to work on the capability that we now call the Geran-2, their version of the Shahed. The idea was that these should be a complement to the more exquisite, expensive, difficult-to-produce weapons. That this country should be producing these lower-end, precise mass systems for itself. The LUCAS, in some ways, was emblematic of that category, but not necessarily the only plausible example.
Q: What was your argument for using these weapons and what was the pushback, if any, and if so, from who?
A: I don’t think that there was pushback. Nobody looked at this and said ‘Oh, that’s a terrible idea.’ But this is just not something that the United States had done before. We were starting to see growing interest in this kind of thing, conceptually, through something like the Replicator initiative, which my office had also promoted. We were trying to push for funding and capabilities, but it was just a very different kind of thing. It wasn’t a ‘let’s produce the best version of something.’ It was ‘let’s invest in a very much lower-cost capability where the idea is that the quantity has quality of its own.’
Q: What did you think when you first saw the LUCAS concept?
A: I thought that this was exactly the kind of system that we had been looking for in a world where defense procurement needed to become more risk-acceptant and not let the perfect be the enemy of the good, given all of the well-known munitions shortages. Even the idea of reverse-engineering a Shahed seemed like an obviously good idea, in some ways, Iran had already perfected. What could we do?
Q: The Wall Street Journal reported that the LUCAS was a mock-up at the time it was selected over other more mature systems. Can you say why?
A: I don’t know the answer to that. As the DASD for Force Development and Emerging Capabilities, my job was to try to figure out, given the different strategies, what were capabilities that we should be pursuing. So something like the LUCAS was right in the wheelhouse that my office was looking for and trying to move money toward, but the specifics of the vendor and how all that was chosen, is a question for Research and Engineering (R&E).
Q: Who did you have to try to convince to pursue this?
A: The amount of money involved was really small given the scale of the Pentagon budget. But because it wasn’t anything that anybody had planned to spend money on, it involved having to go to [Pentagon] elements like R&E and the Defense Innovation Unit (DIU), and going to the military services, and making the case that this was an important capability that the joint force needed, and that could potentially be developed very quickly.
Q: How much money are you talking about? Millions? Tens of millions? Hundreds?
A: To the best of my recollection, it was tens of millions.
Q: Who in the Pentagon made the decision to push this forward while you were there?
A: Eventually, we successfully persuaded then-Deputy Secretary of Defense Kathleen Hicks that, especially at the price point, this was a risk worth taking.
By the way, just to be clear about one thing, [Deputy Assistant Secretary of Defense for Prototyping and Experimentation] Alex Lovett is really the hero of this story in many ways. He both discovered the concept and moved it forward even after all of the original team that supported the LUCAS had left the government. He did incredible work in R&E in making the LUCAS a reality. The people who really deserve the credit are Alex, the CENTCOM team that accelerated it and Arizona-based SpektreWorks, the company that made it.
Q: When the Biden administration left office, where was the LUCAS program?
A: It was moving forward, but not yet a fielded capability. The Trump team deserves credit for seeing the potential and moving this forward and getting it done.
Q: Why has the U.S. been so slow to adopt these types of weapons?
A: Because I think that the American military since the end of the Cold War was built on having the best capabilities. And the idea was that even if we have small numbers of systems, our systems are so good that it doesn’t matter. And it took a while to get the services out of that [mindset] and for people to embrace the idea that second-best military capabilities might have real value. Especially given the very real risk of major conflict.
Q: How could weapons like LUCAS fill in the U.S. magazine depth?
A: I think the math would suggest that building 400 Tomahawks would give you 46,000 LUCAS rounds. And so if you think about this range of capabilities that includes something like LUCAS and like other really super low-cost systems – up through the like family of affordable mass missiles that the Air Force is working on – you’re just talking about like a whole new level of depth in the magazine for the American military.
This is absolutely necessary for the U.S. to continue to compete successfully in the Indo-Pacific in particular, but also in general, as we’re seeing in the Iran context. But the system had been designed for so long to only procure small numbers of the best systems, and it was challenging to [change that]. There’s now a lot of momentum behind that, but it was challenging to get things moving in the other direction.
Q: What was your reaction when CENTCOM first announced that it stood up a LUCAS drone unit and again, when they were first used in combat during Epic Fury.
A: CENTCOM has been on the leading edge of experimentation for years. I saw that firsthand when I was in the Pentagon and so I was not surprised that CENTCOM was leading the way in experimenting with and figuring out what the concepts for use would be for something like the LUCAS and I was not surprised to see CENTCOM figuring out how to apply it successfully in Epic Fury.
Q: What were your thoughts when they were first used in combat?
A: It was a great illustration of how the Pentagon can move fast in developing and fielding emerging military capabilities when it chooses to, but the incentives just haven’t been aligned to do it that way for decades.
Q: CENTCOM commander Adm. Brad Cooper told me LUCAS drones were indispensable. How much do you know about how they were used in Epic Fury, and what are some of the target sets that you think they should be used against?
A: I genuinely don’t know the answer to that. I think someone inside the conflict would be in a better position to answer that.
Q: What would your recommendation be about how they should be used and against what kinds of targets?
A: These weapons are accurate. They’re just pretty easy to shoot down. So in some ways, you could either imagine using these against military targets in large volumes to try to overwhelm defenses, or you could use them in combination with more exquisite weapons to try to trick adversarial defenses and clear the path, in some ways, for more exquisite weapons.
Q: The Russians are doing that in Ukraine, both to overwhelm systems, but they’ve been a very effective strike weapon. If you were still at the Pentagon, how would you recommend they be used in a campaign like Epic Fury and across the military as a whole. And how widely should they be fielded, and what would be ideal numbers?
A: It would not be for somebody like me to decide given my role. But these are essentially inexpensive precision weapons, so any target you would be comfortable using a precision weapon against, in theory, you could use LUCAS against. What’s important is that any weapon has been through the testing and evaluation cycle so you can prove that it works effectively and reliably.
Q: Images released by CENTCOM show what appears to be a gimbaled camera system mounted on its nose and, most importantly, a miniature beyond-line-of-sight satellite datalink mounted on the spine of some LUCAS drones. Being able to be controlled dynamically after launch at great distances, do you see these as a strike weapon to hit moving targets, targets of opportunity and operate in swarms?
A: The missions you could use a weapon like LUCAS for is a software question as much as a hardware question. There’s no reason in principle it couldn’t be used to hit moving targets, or in coordination.
Q: How widely should they be fielded by the U.S. military? What are ideal numbers?
A: I think incredibly widely. This is whether it is LUCAS specifically, or related precision systems. I think if you are looking for the trade space to increase defense production short term – over a couple of years period, as opposed to a decade – are in these lower-cost capabilities that you can scale through more commercial manufacturing. So I would think that there’s an opportunity to have tens of thousands to hundreds of thousands of these kinds of systems to complement exquisite capabilities like Tomahawk.
Q: With a far smaller warhead, they wouldn’t replace Tomahawk, though, right?
A: The LUCAS is a much smaller weapon than the Tomahawk, among other things. It’s not a replacement. What we need is a high-low mix with both types of systems.
Q: How would they be helpful in a fight against China?
A: If you think about the potential range that LUCAS variants could get to – if you look at the max range of the Geran-2 or the Shahed – and then we could do better. I think that means that you now have new attack vectors and the ability to flood the space with munitions in a way that could substantially complicate Chinese defenses.
Q: The LUCAS drone used by CENTCOM was developed by SpektreWorks in cooperation with the U.S. military. Did you ever work with them?
A: No.
Q: SpektreWorks received a $30 million contract in Fiscal Year 2025 from the Accelerate the Procurement and Fielding of Innovative Technologies (APFIT) program to provide procurement funding for innovative projects that have completed development and are ready to transition into operational use. Is APFIT helpful in speeding up the procurement process for LUCAS-type drones?
A: APFIT is a bridge to production for capabilities that are ready to start scaling but not yet programs of record. So APFIT funding is a starting point not the end point.
Q: What are the bottlenecks for the production of LUCAS drones?
A: I think the supply chain issues for the LUCAS and other precise mass systems are different than those facing AGM-158C Long Range Anti-Ship Missiles (LRASM) OR AGM-158 Joint Air-To-Surface Standoff Missile (JASSM). That you can make them with modified commercial manufacturing means that the constraints are simply lower from a production perspective.
Q: What can the government do to speed up the process of procuring LUCAS-type drones?
A: The government should speed up the process by pursuing a Liberty Ship model. Since the government owns the IP, you can have lots of vendors produce the Lucas simultaneously and then increase orders with the vendors who excel the most.
Contact the author: howard@thewarzone.com