The U.S. Air Force is now aiming to have at least 500 Collaborative Combat Aircraft (CCA) in service by 2032. The service also wants to have an equal number of Massed Modular Aircraft (MMA) drones to succeed its MQ-9 Reapers by that time. In terms of stated CCA force structure plans, this is a substantial step up, with the service having most recently talked about acquiring “over” 150 of these drones by the end of the decade.
A composite image showing the FQ-42 and FQ-44 designs together with their new official nicknames. USAF
“We will have autonomous fighters like the Collaborative Combat Aircraft, or CCA. We intend to have at least 500 of these in service by 2032,” Meink said in his speech. “Our intent is to field 100 MMAs in 2029 at even a lower cost than the CCAs and a fraction of [the] cost of manned aircraft we build today. Then by 2032, 500 of these [MMA] platforms will join our force.”
When it comes to CCAs, “today, the first FQ-42 and FQ-44 increments are already rolling off the assembly line,” Meink added. “They’ll be performing many of the same missions that we do with manned fighters today.”
A row of YFQ-42A prototypes. General Atomics previously gave the company name Dark Merlin to this design. GA-ASI
The Air Force’s stated CCA force structure targets have fluctuated significantly in recent years. In 2023, then-Secretary of the Air Force Frank Kendall outlined plans to acquire at least 1,000 CCAs, if not more, across the life of the entire program. However, Kendall and other Air Force officials subsequently talked about a far smaller initial fleet of between 100 and 150 drones.
“These contracts reaffirm our confidence in the strategic path forward for the program to procure over 150 combat capable CCA by the end of the decade,” Meink himself said in a statement in June when the production orders for the FQ-42 and FQ-44 were announced.
A YFQ-44A prototype, with an inert AIM-120 air-to-air missile under its wing, seen during a test. USAF
The FQ-42 and FQ-44 are also being acquired under the first phase, or Increment 1, of the CCA program. At least one additional increment is planned, and could lead to orders for one or more additional types of CCA. Though it is not immediately clear, it is possible, if not probable that Increment 2 CCAs are factored into the latest 500 drone figure.
TWZ’s Jamie Hunter got a close look at the Skunk Works’ Vectis drone at Air, Space, & Cyber 2026. Photo: Lockheed Martin
“We talk about 500 by 2032, [but] we could actually build more of them than that,” Secretary Meink also said today during a question-and-answer session immediately following his speech. The goal of the CCA program “from the get-go” was “simple aircraft that you could build fast and at a lower cost,” he continued.
“So now we’re in a situation to [sic] where we could actually build more than we can actually afford to buy,” he added.
It is unclear what the unit costs of the FQ-42 or FQ-44 are currently. The Air Force has said previously that it is targeting a price point of around $20 million apiece once full-rate production gets underway.
Beyond any fleet size target, the Air Force is still very much in the process of figuring out how future CCAs will be deployed, launched, recovered, supported, and otherwise operated on a day-to-day basis, as well as employed tactically. During today’s question-and-answer session today, Meink drew attention to a dedicated Experimental Operations Unit (EOU) the service has established to lead the process of laying the groundwork for future CCA operations. The EOU notably led a significant new round of operational testing of YFQ-42 and YFQ-44 prototypes at Creech Air Force Base in Nevada back in July, which TWZ was first to report on. The EOU supported another similar test exercise at Edwards Air Force Base in California in April.
A YFQ-44A at Edwards Air Force Base during a test exercise in April 2026. USAF
“Getting them [CCAs] into these developmental kind of operational units, so they can start experimenting with that – familiarity breeds confidence, right?” Meink said.
The Air Force continues to make it clear that it sees its future CCA fleets as essential for providing capabilities and operational capacity necessary to succeed in future conflicts, especially a potential high-end fight against a near-peer opponent like China.
“Collaborative Combat Aircraft change how we project power and generate mass in highly contested environments,” Air Force Chief of Staff Gen. Ken Wilsbach said in a statement when the FQ-42 and FQ-44 production contracts were announced. “Delivering this capability to our warfighters faster ensures our forces maintain the tactical edge required to deter and, if necessary, defeat any adversary.”
The MMA drones are expected to conduct the same general strike and intelligence, surveillance, and reconnaissance (ISR) mission sets the service’s MQ-9A Reapers perform today. Major Reaper losses in the course of operations against Iran, compounded by significant additional attrition and deliberate divestments in recent years, are key drivers behind the MMA effort, which officially kicked off in July.
A US Air Force MQ-9A Reaper drone. USAF
“As we saw in Epic Fury [the nickname for major combat operations against Iran earlier this year], ISR/strike platforms have been essential. We have used MQ-9 and even now the ULTRA aircraft to great effect,” Meink said during his speech today. The MMA program is “building on these lessons” and “will provide affordable, attritable, long-range strike, and we will be able to field them at scale.”
“MMA is designed to deliver the results our combat commanders need,” Chief of Staff Wilsbach said during his own separate keynote at the Air, Space & Cyber Conference today. “That’s how we stay ahead by moving quickly, investing in new tech, and delivering to our warfighters what they need to win.”
If the Air Force can meet its target of having a fleet of at least 500 MMA drones by 2032, this will be another significant boost to the service’s overall uncrewed capabilities. As of May, the Air Force had 135 MQ-9s left in inventory, and there have been more reported losses since then.
General Atomics, which developed the Reaper, has already put forward a new design called Wildfire as a potential successor. Many other companies are already lining up to submit their own proposals to meet the Air Force’s MMA needs.
Between the latest CCA force structure plans and the new MMA program, the Air Force’s fleets of combat drones are on track to look very different by 2032.
Lockheed Martin’s Skunk Works advanced projects division has shared all new information about its stealthy Collaborative Combat Aircraft (CCA) type drone design named Vectis, which it first unveiled a year ago. One of the most intriguing details is the true nature of its air inlet configuration, pointing to the premium on survivability the Skunk Works has placed on its CCA offering, especially when compared to other CCA designs.
Vectis has a far more exotic, deeply recessed “submerged” top-mounted air intake than was originally disclosed when Vectis first came out of the shadows. This further underscores the heavy emphasis Skunk Works says it has been placing on its CCA’s ability to stay undetected and live to fight another day. Vectis is also the first tailless CCA design we have seen from a U.S. manufacturer.
An updated view of Vectis. Lockheed Martin Skunk Works Lockheed Martin
Ron Fehlen, Lockheed Martin Vice President and General Manager of Skunk Works, gave TWZ and other outlets an update on Vectis last week ahead of today’s start of the Air & Space Forces Association’s 2026 Air, Space & Cyber Conference. Skunk Works says work on constructing the first prototype of Vectis – a word that means lever or pole in Latin, and that Lockheed Martin has said reflects the ‘leverage’ the drone offers – is underway now. The goal is to have it flying by the end of next year. There are already plans to build four more prototypes using internal funding.
Check out the Skunk Works’ new Vectis video below:
Source: Lockheed Martin
“It is designed, as you can likely see, to be both survivable and affordable and flexible across the range of missions that our customers need, have needed, and continue to need into the future,” Fehlen said by way of introduction. “The technical specs, as you can imagine, are designed to meet and, in some cases, exceed the customer sets that we are working with and their requirements and their mission needs.”
The basic core of the single-engine Vectis that we have seen now is largely unchanged from what had been shown at the time of the unveiling last year. It is still a tailless drone that has a lambda wing planform and air intake mounted on top of the central part of the fuselage, as well as a prominent chine line around the forward end and a shovel-like shape to its nose. New renderings do show revised positions for various conformal antennas and/or sensor apertures around the central fuselage. These are all established features of low-observable (stealthy) aircraft designs.
Skunk Works’ Fehlen also shared last week that Vectis is approximately 34 feet long and has a wingspan of around 38 feet. As a comparison, the Navy’s two-seat primary jet trainer, the T-45 Goshawk, is roughly 39 feet long by 30 feet wide.
The biggest new detail, as mentioned, is the shaping and position of Vectis’ air inlet. Fehlen stressed that this is actually not a new feature, and has “been a part of the design since the beginning,” but was not reflected in previously released renderings and scale models.
Another newly released rendering of Vectis showing the submerged intake. Lockheed Martin Skunk Works
The intake was depicted in the past as having a relatively straightforward configuration. It did appear to have a saw-tooth leading edge, another indication of stealthy design considerations. The visuals we have now show the intake is deeply recessed into the top of the fuselage, sitting behind a downward-sloping v-shaped channel.
A rendering of Vectis that Lockheed Martin released last year that did not depict the submerged intake. Lockheed MartinSkunk Works
Vectis doesn’t have a fully flush-mounted inlet, like what is found on Boeing’s MQ-25 Stingray tanker drone, among other designs, but it is close to it, very loosely similar to what’s found on the B-21. The rear is slightly raised above the v-like pocket where air flows down into the intake. Flush or near-flush intakes offer particular benefits when it comes to low-observability (stealthiness) and aerodynamic efficiency, but are also notoriously problematic, especially in certain flight regimes, as you can read more about here. This is even worse for high-performance aircraft that have a wide operating envelope and have to maneuver at higher angles of attack. The CCA would likely fit into this category, at least to a degree.
When it comes to Vectis, “there’s a lot of different reasons behind it [choosing this submerged intake configuration], but I’ll pick up on a couple here for you,” Skunk Works’ Fehlen explained. “The first is it reduces that frontal drag. So, anything sticking out in the wind, as many – as all of you know, I’m sure, creates drag on the aircraft itself. So, if you want better performance from a range, et cetera, then you need to be able to reduce that drag, and that’s an example.”
“It’s also interesting to me, particularly as the Air Force speaks about Agile Combat Employment, the ACE concept, as you’re aware. There is an advantage to having an inlet on the top of the aircraft if you’re out in more austere environments and you’re concerned about FOD [foreign object debris] pickup off of a runway,” he added.
Another newly released rendering offering a different perspective on the submerged intake. Lockheed Martin Skunk Works
Fehlen also confirmed that the inlet’s design has to do with survivability, but did not elaborate in any detail, which is unsurprising considering the sensitivities of such design elements. The original Vectis promotional video Lockheed Martin put out last year, seen below, showed an S-shaped duct behind the air intake and exhaust shrouding, which are features designed to further reduce a platform’s signature.
In terms of performance, Lockheed Martin has yet to share any hard specifications it is targeting for Vectis, or disclose details about the engine the design will use. Fehlen did also highlight the drag-reducing performance advantages of the drone’s tailless design, including for maximizing range.
Details about what other capabilities Vectis might have are still limited. Fehlen noted that the design is expected to be capable of both air-to-air and air-to-ground mission sets. In a new computer-generated promotional video, seen earlier in this story, the drones are depicted launching what look to be AIM-120 Advanced Medium Range Air-to-Air Missiles (AMRAAM) and GBU-53/B Small Diameter Bomb II (SDB II) glide bombs, also known as StormBreakers. The AIM-120 has already emerged as something of a default initial weapon for Western CCA-type drones. The video also confirms Vectis has an internal bay that is able to accommodate at least two munitions, likely the size to accommodate medium-range air-to-air missiles.
Captures from Lockheed Martin’s new Vectis promotional video depicting the drones launching what looks to be an AIM-120 missile, at left, and a pair of GBU-53/B SDB IIs, at right. Lockheed Martin capturesA rendering of Vectis with its elongated internal bay open from the new promotional video. Note the AIM-120 in the background. Lockheed Martin captureAnother screen capture from the new Vectis promotional video showing two munitions loaded inside the drone’s internal bay. Lockheed Martin capture
In addition, the video shows Vectis as having a traditional electro-optical/infrared sensor turret behind a stealthy faceted enclosure under the nose. When the drone was first unveiled last year, TWZ noted that it could carry a variety of sensors, including compact radars and/or infrared search and track (IRST) systems, which would be particularly relevant in the air-to-air role.
The sensor turret inside the stealthy faceted enclosure as seen in the new Vectis promotional video. Lockheed Martin capture
During the press call, Skunk Works’ Fehlen reiterated that Vectis is being designed with an eye toward modularity and open-architecture systems. This, in turn, will make it easier to integrate new and improved capabilities and functionality down the line. Open architecture is at the heart of every CCA offering we are aware of at this time.
Fehlen further highlighted Lockheed Martin’s extensive work to date on advanced autonomy packages and crewed-uncrewed teaming capabilities. This includes the development of the Multi Domain Combat System (MDCX), an “autonomy platform” that is already central to the U.S. Navy’s plans for the carrier-based MQ-25 Stingray. The head of Skunk Works also explicitly called attention to autonomous air intercept testing conducted in cooperation with the Air Force earlier this year using the unique X-62A test jet, and said that work is feeding into Vectis’ development.
The X-62A test jet seen during the autonomous air intercept testing earlier this year. USAF
On top of all this, Lockheed Martin is already actively laying the groundwork for the U.S. Air Force’s F-22 and F-35 fighters to control future CCA drones, via in-cockpit touchscreen-type user interfaces, likely involving tablets, at least at first.
The new Vectis promotional video also shows a traditional ground control station being used to direct the drones, but with the phrase “managed from anywhere” on the screens, as seen below. Fehlen did say that any specific requirements for autonomy, command and control, and other operational capabilities would be up to individual customers, and that third-party vendors could be involved.
Lockheed Martin
Beyond Vectis’ design and expected capacities, Fehlen touted the speed at which work on the first physical prototype is progressing, and highlighted the use of digital design techniques and advanced manufacturing processes.
“This team, when kicked off, took a design on a napkin, which we will likely frame one day in my office … two months later, they were in the wind tunnel validating the design,” he said. “So, in under two months, we’re already getting in the wind tunnel straight from off of a napkin. We had the first digital build in six months, and the first part release in eight months. So it gives you a touch of a sense for exactly how fast we’re moving.”
“This aircraft was designed in a fully integrated digital environment,” he added. “Really, it’s creating that unbroken digital thread from the design all the way through sustainment.”
“We’ve used advanced design and manufacturing processes developed for many years under our own internal investment … for other Skunk Works programs, we’ve pulled that into this, of course, to continue to advance our manufacturing as well as the advanced design capability,” Fehlen continued. “They’re using real production processes.”
A central fuselage part of the first Vectis prototype now under construction. Lockheed Martin Skunk Works
The Skunk Works boss stressed that this approach offers advantages not just for accelerating development and prototype, but also for reducing costs and positioning Vectis to enter full-scale production.
“It gives you the reliance and the confidence as you continue to move forward that you’re ready to off and build these things because you’re using those real production processes,” he explained. “An example of that is a technique that we call minimal tooling determinant assembly. It’s a technique that really enables an increased level of affordability and speed through really reducing the amount of touch labor required and increasing the first pass yield. We’ve seen, in many cases, an 80 percent reduction in labor hours to build up the aircraft itself.”
Part of one of the wings for the first Vectis prototype. Lockheed Martin Skunk Works
Fehlen made an indirect reference to flat-pack furniture from IKEA to provide additional context, in extremely general layman’s terms, about what this technique offers.
“That’s what we’re talking about, that idea that it can come together very quickly, limited touch, and limited error. We constrain that to reduce the number of labor hours, which is not just a cost implication, but also buoys the speed and how fast we’re going to be able to produce them,” he said. “You can imagine, as well, there’s a lot of simplification that goes into this. We have significantly reduced the part count. Again, back to simplify the build so that you can have a positive impact on speed and affordability, not new techniques, but we’ve applied them here with vigor to ensure that we are getting to the cost point that the services and our customers need in order to make this system effective.”
Another element of the first Vectis prototype, which is now being built. Lockheed Martin Skunk Works
In terms of actual unit cost, “we’re always looking at, and, of course, we’re targeting to be [at] the CCA competitive price, so we’re very cognizant of it and make trades associated with that,” according to Fehlen. Vectis is “focused more on the CCA side of the house, and obviously the price points that are highly relevant in that regime, and then, of course, making trades along the way, and to make sure that we can meet it.”
The head of Skunk Works was responding here specifically to a question that mentioned the Air Force’s target of a roughly $20 million unit price for drones acquired under the first phase, or Increment 1, of its CCA program once full-scale production is achieved. The service announced in June that it was ordering fleets of both of the Increment 1 designs now in development, General Atomics’ FQ-42A Dark Merlin and Anduril’s FQ-44A Fury. The Air Force has plans for at least one more increment to this program, and has said in the past that it could pursue lower-cost (and less exquisite) designs.
A trio of YFQ-42A Dark Merlin prototypes. GA-ASIA YFQ-44A Fury prototype with an inert AIM-120 missile under its wing. USAF
“While not discussing suppliers, you can imagine we are ensuring that all of the decisions we make in the development and then the build out of these aircraft are designed around achieving that competitive cost point, while at the same time maximizing the amount of capability,” Fehlen added. “Those are the things that our customers are looking for, and we put it in a survivable solution.”
“There are some [companies] that are really going down the attritable [route], which defines a different price point than the CCA community has been targeting,” he also noted. Attritable designs are “quite a bit less from a cost point perspective and often from a capability perspective.”
With this in mind, there are still larger questions about where exactly Vectis may fit into U.S. military CCA plans, as well as those emerging from allied and partner air arms globally. Skunk Works has said in the past that it was eliminated from the Air Force’s CCA Increment 1 competition because it put forward a very stealthy “gold-plated” design that was not what the service wanted. Lockheed Martin had also indicated that its view of the capability/cost balance on new CCA-type designs had subsequently shifted very much in the opposite direction. Though it is still not entirely clear, what we’ve learned about Vectis in the past year would seem to signal that the pendulum at Skunk Works has swung back the other way again, at least to a degree.
As an interesting point of comparison, Northrop Grumman had also entered a higher-end design for the Increment 1 competition. In the wake of that loss, the company produced a much more familiar CCA design with a more traditional planform, which is now known as the YFQ-48A Talon Blue.
It’s also worth noting here that Skunk Works had publicly shared a future crewed-uncrewed teaming vision that included multiple tiers of drones back in 2022. This ranged from a cruise missile-like design the company now calls the Common Multi-Mission Truck (CMMT) all the way up to a concept for what could have been a very low-observable (stealthy) and highly autonomous flying wing uncrewed combat air vehicle (UCAV).
“We listen closely to our customers to deeply understand the operational environment and use those insights and our own analysis to inform the diverse solutions we develop for different mission sets,” Lockheed Martin told TWZ directly after the press call last week in response to a follow-up question about whether the 2022 vision was still relevant and if Vectis had supplanted the upper-tier concept shown at that time. “We continue to evaluate a range of crewed and uncrewed concepts against evolving customer needs.”
“Lockheed Martin Vectis addresses a warfighter need for a survivable, flexible and multi-mission capability, both domestically and internationally,” the company added in its statement. “We’re leaning into that demand, making strategic investments to accelerate Vectis’ progress and maturity and to deliver the capabilities our nation and allies need.”
A screen capture from a more recent Lockheed Martin video showing renderings of various different drone designs, including Vectis (at lower left), the CMMT (at upper right), and the stealthy flying-wing (at lower right). Lockheed Martin capture
Lockheed Martin and a host of other companies globally are being presented with a clearly expanding CCA market, but there is also still obvious interest in higher-tier UCAVs. China, in particular, has positioned itself as the current leader in the stealthy flying wing UCAV space, as well as a dominant actor in advanced combat drone developments more generally speaking. Russia, India, Turkey, and France continue to pursue flying wing UCAVs, as well. That being said, Skunk Works’ Vectis is arguably the closest we have seen to a low-observable UCAV to come out of recent CCA developments.
Regardless, Skunk Works is confident that there is a place for Vectis, together with autonomy and other capabilities it is developing, domestically and internationally, in this future uncrewed air combat ecosystem. It “is a survivable autonomous multi-mission platform, which we believe is the solution needed within today’s operational environment and into the future,” Fehlen said last week.
Skunk Works is committed to “delivering what’s next across … airframes, open architectures, to include artificial intelligence and autonomy, and then crewed-uncrewed integration,” he added. “Make no mistake, it will take advancements across all these areas to get to where our customers are going. We look at the full picture. Vectis is a holistic weapon system from our perspective.”
Altogether, “it’s an urgent need. We know that. We can hear it.”
UPDATE: 3:10 pm EDT
We now have a photo of the full sized Vectis mockup on the floor or Air, Space, Cyber 2026:
La superficie forestal quemada en lo que va de año en España por incendios asciende a 296.502 hectáreas, lo que supone alrededor de 3,5 veces más que la media anual desde 2006, según datos actualizados del Sistema Europeo de Información Sobre Incendios Forestales (EFFIS), hasta el 9 de septiembre.
Earlier this year, an Iran-linked actor used Anthropic’s Claude artificial intelligence model to scrape and analyze data with the aim of helping target U.S. naval forces in the Middle East. A group in Yemen – very likely Iranian-backed Houthi militants – also used Claude to assist with work on guidance systems for ballistic and hypersonic missiles.
These are just some of the revelations about how adversaries and other malign actors have been using one of the premier pieces of AI software being developed in the United States that were contained in a new report Anthropic released late yesterday. The 154-page document also details instances where Claude was used directly and indirectly to support intelligence gathering and surveillance, cyberattacks, influence operations, and even worrisome biological research over the past eight months. In addition to Iran and Yemen, the human actors behind these activities were based in Russia, China, and a host of other countries.
Anthropic has shared details about what it calls the “misuse” of its models in the past, but describes the new report as its “most detailed” to date. The company has clearly been cataloging these case studies, and has assigned each one a control number. Though the entire contents of the report are cause for concern, the explicit attempt to use Claude to target American forces, as well as to develop new higher-end missiles and other military technology, are particular standouts and signs of what’s to come, as we will highlight below.
GTG-30005: Military reconnaissance
GTG-30005 is the case study dealing with the targeting of U.S. naval forces, and the summary is as follows:
“In another investigation, we identified and disrupted an Iran-nexus threat actor that used Claude to collect and analyze publicly accessible data to develop targeting recommendations against US naval forces in the region. The threat actor used Claude to compile targeting handbooks, through a Python pipeline the threat actor built with Claude’s assistance, to identify and track naval positions based on open-source information. The compiled material included a roster of US personnel scraped from captions on public military photographs; publicly accessible ship and aircraft transponder identifiers; commercial satellite-imagery query scripts; and an inventory of public websites that exposed US naval movements. The threat actor also directed Claude to compile vulnerability research on shipboard systems, cataloging known CVEs in maritime VSAT terminals, Cisco communications equipment, and industrial control products.”
“We banned the actor’s account, developed detections to reduce the risk of future misuse, and shared threat intelligence with government authorities to disrupt the threat.”
The Nimitz class aircraft carrier USS George H.W. Bush and several other US Navy warships sail together in the Arabian Sea in June 2026. CENTCOM
How exactly Anthropic defines the term “nexus” here is not immediately clear, and other entries in its new report also use “state-nexus” in addition to country-based variations. However, in this specific case, an addendum is added to the entry that explicitly says that the same account was separately tied to “domestic mass surveillance” software development work “for Iranian state systems,” clearly pointing to an actor directly linked to the regime in Tehran.
The use of AI technology by nation-state armed forces to develop target sets and engage them is already known to be growing globally. The Israel Defense Forces (IDF) have been a pioneer in this regard, but the U.S. military, China’s People’s Liberation Army (PLA), and others are also known to be developing and fielding these capabilities.
GTG-87001: Disrupting a Yemen-based guided weapons engineering cell using Claude to develop guidance software
Anthropic’s new report details an instance where an individual in Yemen used Claude as part of missile guidance work, which is given the control number GTG-87001.
“We identified a cell of threat actors based in northern Yemen running three weapons development programs: a guided rocket that used a commodity phone-class flight computer with final-phase homing guidance; a multi-stage ballistic missile with a stated range goal above 2,000 km; and a multi-variant missile (referred to as the “R2000” set) that included a hypersonic glide vehicle variant.”
“The actors used Claude Code in place of human software engineers to develop the guidance, navigation, and control (GNC) software that steers and stabilizes a flying vehicle. For example, they used Claude to integrate an open-source autopilot onto a phone-class flight computer, writing the control and position estimation software, tuning the control settings, running a firmware build pipeline, and performing a flight simulation. The actors managed several Claude instances at once, assigning each one a role, much as a lead would delegate work on a small engineering team: the actors tasked one instance with writing the code, another with research, and a third with reviewing the code the first instance produced.”
A graphic included in Anthropic’s new report summarizing the ways in which Claude was used by the actors in Yemen to support missile guidance system development. Anthropic
“Our safeguards blocked many of their requests, but not all of them. The actors used a variety of tactics to evade our safeguards, including hiding their goals and the products the software was meant for, and they split their work across multiple sessions so no single session revealed their full intent.”
“These actors carried out a sustained effort to develop guided weapons, including using Claude to design guidance software. We do not have evidence the actors succeeded in fielding an operational device; but they did test-fire a guided rocket. This field test appears to have failed: within hours, the actors returned to Claude to work out why it failed.”
“We identified this activity as part of our internal investigations into suspected weapons development. We banned accounts associated with the actors and shared threat information with public- and private-sector partners to mitigate risks posed by the actors. Nevertheless, we have evidence that the actors had already built an offline simulation toolkit that does not rely on Claude or other engineering computing environments such as MATLAB.”
As mentioned earlier, the report does not name the Houthis, but that group has historically been based primarily in the northern end of Yemen. It is also the only entity in the country known to have active missile development programs, which are conducted with critical support from Iran.
Whether this particular work directly resulted in any new capabilities is unknown. Anthropic notes that there looked to be at least one failed test launch before the accounts were banned, which would have occurred sometime after the beginning of this year, based on the stated cutoff dates for the report. By the end of 2025, the Houthis had already amassed a wide array of very real ballistic missiles, including types with at least claimed ranges of 2,000 kilometers (approximately 1,242 miles). The group also has a variety of cruise missiles and one-way attack drones that it has employed operationally.
An example of just one of the many types of ballistic missiles in the Houthis’ arsenal. Houthi-controlled media
Still, being able to leverage Claude could only have helped any domestic missile development efforts in Yemen, which might help reduce dependency on Iran, at least to a degree. Furthermore, developing functional hypersonic glide vehicles capable of prolonged stable flight has historically been notoriously difficult, even for entities with extensive relevant knowledge bases on an organizational level.
GTG-17001: Disrupting a China-based operation using Claude to draft a fire control specification and acquisition documents for undersea warfare
Use of Claude in foreign military endeavours goes beyond the Middle East, as evidenced by case study GTG-17001. This involved a Chinese actor conducting work related to a naval counter-torpedo defense system.
“We identified a China-based threat actor who used Claude to advance three parallel tracks of work on an anti-torpedo weapons system:”
“First, the actor used Claude to draft a Chinese-language specification for an anti-torpedo fire control system (the core logic that aims and times an anti-torpedo weapon’s response). The document was written to win approval from a Chinese defense manufacturer, which would move the work on to technical certification and operational testing.”
“Second, the actor used Claude to produce a Chinese-language technical proposal of more than 200 pages, accompanied by an executive briefing deck.”
“Third, the actor used Claude to benchmark their own system against specific US anti-torpedo and anti-submarine programs based on publicly accessible information. They then generated a Chinese-language briefing on US Navy systems derived from open-source reporting.”
“The actor presented themselves as an original equipment manufacturer in the US defense sector. We assess the actor was associated with a Chinese defense industry manufacturer aiming to produce a weapons specification and acquisition proposal for the People’s Liberation Army Navy.”
This picture shows a test of an anti-torpedo system fitted experimentally to a Nimitz class aircraft carrier in the 2010s. USN
“The actor used Claude to write the acquisition proposal, refining it over many drafts. After each draft, the actor instructed Claude to role-play a hostile expert reviewer to critique the proposal, then used that feedback to sharpen the next version. In parallel, the actor used Claude to build pieces of the anti-torpedo weapons system’s fire control software and a test matrix to validate them.”
“The operational lift the actor achieved was a function of using Claude to automate complex technical outputs. The actor leveraged the model to compress the development timelines for the certification registry, compliance documentation, and automated fire control logic. The actor also accelerated the traditional human review cycle by having Claude critique the acquisition proposal across multiple rounds of review while role-playing a persona.”
“We uncovered this activity as part of our internal investigations into suspected weapons development. We cannot attribute the activity to a specific entity or actor. But we have banned the account for violating our Supported Regions Policy and our Usage Policy, which prohibits weapons design and development, and incorporated our investigative findings into our safeguards to mitigate the risk of future misuse.”
Anthropic’s report does not name the U.S. programs leveraged here, but does note the information in question was publicly available. Anti-torpedo systems for surface warships and submarines have been an area of significant interest for the U.S. Navy for years now. Those developments have also produced compact torpedo designs for use in other applications.
A Very Lightweight Torpedo (VLT), a compact torpedo design developed by Northrop Grumman. Northrop Grumman
The PLAN’s interest in these same kinds of capabilities is not at all surprising. Underwater threats and countermeasures to them, broadly speaking, would play a central and vital role in any future conflict between the United States and China across the broad expanses of the Pacific.
GTG-27005: Disrupting a Russia-based operation using Claude to engineer an autonomous military drone swarm
Another case study, GTG-27005, details the use of Claude by Russian actors working on drone swarming technology.
“We identified likely freelance Russia-based threat actors who set out to build a full-stack autonomous first-person-view (FPV) kamikaze drone swarm. The actors used Claude Code to write and test the code and save it directly into the actors’ own project files. In addition to Claude Code, the actors used a software-in-the-loop simulation stack and a rented graphics processing host for model training. They called the operation ‘DronDoc’ or ‘Serafim.'”
“The actors used Claude to build the core software system, including the drones’ shared swarm memory and fault-tolerant coordination logic (FTCL); an onboard small language model to govern attack, observe, and return-to-base behaviors; a terminal guidance software system to steer drones to their target (using the onboard camera) and issue the call to detonate; a control-link geolocation module to find opposing drone operators; a passive acoustic detection layer; and low-level logic for the drones’ programmable chips. The actors designed the platform for autonomous lethal engagement; the onboard model could select targets (including a ‘person’ target class) and issue detonation commands without a human in the loop. The actors’ activity—including flashing the low-level firmware to live development boards, provisioning single-board computers, and wiring up a simulation environment over a mesh network—confirmed that they were using real hardware-in-loop testing within their sessions.”
Many were surprised yesterday by the news that a Russian fiber-optic FPV drone flew into Kramatorsk and attacked a car.
But there is nothing surprising here. The war of 2025 is already very different from the war of 2024. From LBZ to Kram — 20 kilometers. Enemy FPVs can fly even… pic.twitter.com/hTfhJFPcxZ
“The actors trained a computer vision classifier on scraped Ukrainian combat footage, splitting the target classes into “enemy” and “friendly,” and allow-listing Russian systems. They also repeatedly used a fixed coordinate in Donetsk Oblast as the demonstration strike point, with front-line cities and corridors in Ukraine as the mission geography.”
“The actors created their accounts between late 2025 and early 2026 and started the operation in mid-May 2026. The actors circumvented our geographic access controls by routing traffic through commercial virtual private servers.”
“We assess the actors were a small, specialized freelance team doing a mix of civilian and military work, not a Russian state entity. We identified nine accounts associated with this group; eight were used only for ordinary freelance work, not weapons-related software development. Based on our investigation, we assess the actors had ties to a regional university with a federal research center associated with the Russian Academy of Sciences. The actors claimed to have received funding from Russia’s Advanced Research Foundation, National Technology Initiative, and Ministry of Defence, though we cannot verify those claims. We identified this activity as part of our internal investigations into suspected weapons development, we banned accounts associated with the actors, and have incorporated our investigative findings into safeguards to reduce the risk of future misuse.”
Anthropic’s report included a table, reproduced below, detailing different types of drones and other capabilities that were fed into the model as part of this work.
GTG-17002: Disrupting a China-based operation using Claude to build targeting software for electronic warfare and air defense suppression
The last case study in the Conventional Weapons section of Anthropic’s new report, GTG-17002, covers electronic warfare-related work by a Chinese actor, with a particular focus on potential air defense targets in Taiwan.
“We identified a China-based actor who used Claude’s chat, coding, and agentic work tools to design, build, and iterate on a Chinese-language suite of about 16 modules for electronic warfare, using the electromagnetic spectrum to detect, jam, or deceive an opponent’s radar and communications, and for suppressing an opponent’s air defenses.”
“The actor used Claude to build the software system, from the underlying logic to the user interface, and iterated through 12 versions. This included implementing and optimizing the system’s radar detection and jamming physics, generating a vulnerability analysis module, and drafting Chinese-language targeting instructions. The software suite the actor built analyzed an opponent’s radars, surface-to-air missile sites, command posts, and communications nodes, then computed their detection coverage, assessed the effectiveness of jamming, ranked targets by value and vulnerability, including which to suppress first, and determined how best to assign jammer sorties to targets across multi-day campaigns. The suite also ranked which of an opponent’s assets to suppress first, and modeled specific engagement envelopes, including those of Patriot and THAAD-class [Terminal High Altitude Area Defense] systems.”
“Mid-project, we observed the actor change the simulation’s default scenario to 12 targets in Taiwan. The targets included a command bunker in Taiwan, an early warning radar site, Patriot and Tien Kung [surface-to-air missile] batteries, major air bases, and a regional combatant command headquarters.”
“The actor also ran a self-hosted model on an internal network alongside Claude and connected the software suite to this model through a tool-use integration.”
“Based on our investigation, we assess the actor is a China-based defense and military-industrial researcher. Account-level metadata and content flagged by our safeguards indicated the actor was linked to PRC research institutions, including the PLA Academy of Military Sciences. We detected this activity as part of our internal investigations into suspected weapons development, we banned accounts linked to the actor, and have incorporated our investigative findings into our safeguards to reduce the risk of future misuse.”
An accompanying table, seen below, mentions several specific types of U.S.-made air defense radars, as well as the Link-16 datalink network, as having come up in the course of this work. Electronic warfare variants of the Chinese Y-8 and Y-9 aircraft, as well as the J-16D electronic warfare jet, and what appears to be a reference to the Golden Eagle drone helicopter, are also mentioned.
A Chinese Y-8GX-11, one of several electronic warfare variants of the Y-8/Y-9 design. Chinese internet via X
The suppression of enemy air defenses is an essential part of any modern air campaign, whether conducted independently or as a part of a larger operation. This would be no less critical in any military intervention against Taiwan from the mainland, or any other larger-scale conflict the PLA might find itself in.
Guardrails and ethics
Beyond the details in any of the specific case studies, Anthropic’s new report underscores growing broader concerns about guardrails for the use of AI and general ethics surrounding the technology. As noted in the instances above, Anthropic has also implemented various safeguards to try to block access to Claude from certain regions and/or to prevent the model from conducting certain types of work. At the same time, what the company has now shared makes clear that those protections can still be circumvented to a significant degree, at least for now.
TWZ has noted in the past that America’s adversaries are likely to be less worried about ethics, in general, when it comes to AI. Anthropic’s new report also shows there are few bounds to what state-aligned or even non-state actors might seek to use this technology for, including the development of potential weapons of mass destruction.
At the same time, AI models like Claude are not going away, and developments continue to advance at an ever-quickening pace, opening the door to further proliferation. Developments are rapidly intermingling globally, with companies in China notably leveraging models created in the U.S. to further their own work, which is then released publicly. The technology will increasingly become more integrated and critical to most aspects of military operations and weapons development, alongside its commercial and civilian applications. It is just as much a superweapon as anything else, and the country with the more capable models, and the willingness to deploy them, will have a major edge in future conflicts.
With all this in mind, it seems very plausible, if not probable, that the case studies described in Anthropic’s new report are just the tip of an approaching iceberg that will need a concerted effort to tackle.
Lockheed Martin and a team made up of Boeing and Anduril are now facing off head-to-head to supply a new middle-tier surface-to-air missile to the U.S. Army. The service wants a second interceptor for its new Enduring Shield system, which can already fire AIM-9X Sidewinders, to expand its capabilities, especially against incoming cruise missiles.
Boeing and Anduril announced their offering had been selected to move ahead to the next phase of the competition to develop what is formally called the Integrated Fires Protection Capability Increment 2 (IFPC Inc 2) Second Interceptor on Wednesday. The Army subsequently confirmed to Breaking Defense that Lockheed Martin advanced to phase 2 of this program. Rafael Systems Global Sustainment, a U.S. subsidiary of Israeli defense contractor Rafael, had also been in the running, but was not down-selected.
The Army’s search for a second interceptor for Enduring Shield traces back to at least 2023. The system, developed by Leidos’ subsidiary Dynetics, is entering operational service now paired with the AIM-9X. It is designed to provide short-to-medium range defense against a variety of aerial threats, including drones. The Army has not had any kind of mid-range air defense capability since it retired its Hawk surface-to-air missile systems in the 1990s.
A US Army Enduring Shield launcher seen deployed in South Korea. USAF Pfc. Sang Woo Choi
A model of the missile Boeing and Anduril have put forward was only shown publicly for the first time earlier this year. Boeing says it is a clean-sheet design and it uses a solid-fuel rocket motor provided by Anduril.
A model of the Boeing/Anduril interceptor seen on display at the 2026 Space & Missile Defense Symposium in August of this year. Jamie Hunter
In its announcement about the Army’s down-selection, Boeing shared a previously unreleased rendering of the design, which notably had two sets of fins toward the tail end of the body. The model that has been shown so far only has one set of fins at the very rear of the body. Whether the rendering reflects a newer or older version of the design is unclear. What is seen in the rendering may also be an additional booster rocket motor not seen on the model, which could be necessary for ground-launched applications.
The rendering that Boeing released this week as part of the announcement of its selection to proceed in the IFPC Inc Second Interceptor competition. Boeing
Boeing has also said that the new interceptor is being developed with an emphasis on lower cost and producibility. Modularity and open-architecture systems to enable “faster, lower-cost upgrades by allowing seeker, avionics or software blocks to be swapped without redesigning the entire weapon” have also been particular focus areas, according to the company.
Other details about the missile’s design and capabilities, including how it is guided, remain limited.
“Our scalable production approach is designed to provide a reliable, affordable layer of defense against a range of aerial threats to protect the warfighter with practical, dependable capability when it matters most,” Bob Ciesla, vice president of Boeing Precision Engagement Systems, had said in a statement earlier this week.
Lockheed Martin first showed a rendering of its design last year, but has also shared relatively specific details about it since then. For what has been seen so far, it is visually reminiscent, at least in broad strokes, of a compact air-to-air missile concept called Cuda, which the company previously promoted for a time in the 2010s.
A rendering depicting Lockheed Martin’s proposed interceptor being fired from Enduring Shield launchers. Lockheed Martin
The design is “not new,” Chris Murphy, senior manager of Business Development for Integrated Air and Missile Defense Advanced Programs at Lockheed Martin’s Missiles and Fire Control division, confirmed to TWZ in October 2025.
“There are technologies that we already have that we’re leveraging,” Murphy added at the time. “But the requirements for the missile are such that they’re going to be new technologies, and there’s going to be some level of development in order to get those integrated.”
“So you know that our [PAC-3] MSE missile [for the Patriot surface-to-air missile system] is a hit-to-kill system, right? We have great confidence in that technology,” he also said when asked specifically about whether the interceptor would have a traditional warhead. “We also work [on] blast-frag[mentation warheads] on other programs. So we will look at everything. That’s part of the trade space the Army has asked us to look at.”
The PAC-3 Missile Segment Enhancement (MSE) is generally presented as a hit-to-kill weapon, meaning that it is designed to destroy its target through sheer force of impact. However, it is actually something of a hybrid design that includes a high-explosive “lethality enhancer” charge to increase its chance of scoring a hit, something you can learn more about here.
A breakdown of the PAC-3 MSE’s key features, including the “lethality enhancer.” Lockheed Martin
For its part, the Army has talked in the past about wanting its second interceptor for Enduring Shield to offer roughly comparable capabilities to the AIM-120D Advanced Medium Range Air-to-Air Missile (AMRAAM), but in the same form factor as the AIM-9X. A key driving factor behind the size requirement is retaining magazine depth. The system’s existing palletized launchers can currently be loaded with up to 18 Sidewinders in individual canisters.
“I do want to highlight a couple of things that I think are very critical gaps that we have. I don’t think – I cannot add anything to the emphasis on magazine depth for currently fielded systems, or for the need for cheaper and [a] wide-ranging mix of interceptors,” Lt. Gen. John Rafferty, head of the Army’s Space and Missile Defense Command (SMDC), told TWZ and other attendees in a keynote address at the 2026 Space & Missile Defense Symposium last month. “That couldn’t be made more loud and clear from the highest levels of our government.”
At that time, Rafferty also stressed the need to further expand the Army’s ability to provide multi-layered air and missile defense capabilities. Enduring Shield, as the Army’s first middle-tier air defense system in decades, is already a key component of that broader effort.
The Army also more specifically sees Enduring Shield as central to reducing over-reliance on its Patriot surface-to-air missile systems. The immense strain on the service’s Patriot force and its inadequacy to meet existing operational demands are well established at this point, and have only been reinforced by the latest conflict with Iran. The Army is notably in the process of establishing its first new Patriot battalion in years, on Guam, which will actually be a composite unit also equipped with Enduring Shield. Pairing the two systems together helps free up Patriot batteries to engage higher-tier threats, especially ballistic missiles, which may themselves be targeting air defenders directly.
The Army has now at least moved to the next phase of the competition for a second interceptor for the Enduring Shield system, demand for which is already growing.
Work at a specialized underground weapons test site in New Mexico, contracted earlier this year, was linked to planning for a potential new operation targeting Iranian nuclear facilities, according to a report from CNN. The U.S. military has notably left Iran’s hardest-to-reach site under Pickaxe Mountain untouched so far, but President Donald Trump and other U.S. officials have been openly talking about striking it.
On February 27, the day before U.S. and Israeli forces launched their initial wave of strikes on Iran, the Pentagon’s Defense Threat Reduction Agency (DTRA) quietly announced it had awarded a sole-source deal to Pate Construction Co., Inc., for new construction at the so-called Alternate Seismic Hardrock in Situ Test (ALT-SHIST) site. The work was needed to support an upcoming classified live-fire test. ALT-SHIST is situated within the White Sands Missile Range (WSMR), a sprawling U.S. Army-managed complex in New Mexico’s Jornada del Muerto desert basin. DTRA is a relatively small and obscure organization that plays a leading role in U.S. military efforts to develop new and improved ways to respond to Weapons of Mass Destruction (WMD) threats.
WSMR and DTRA have come up in discussions about strikes on Iran in the past. The U.S. military’s heaviest conventional air-dropped weapon, the 30,000-pound-class GBU-57/B Massive Ordnance Penetrator (MOP) bunker buster bomb, has been extensively tested at multiple sites there in the past. DTRA was also directly involved in its initial development. MOPs made their combat debut during U.S. strikes last year on underground Iranian nuclear facilities at Fordow and Natanz as part of Operation Midnight Hammer. The possibility has been raised that a different site within WSMR may have been used to test the plan of attack against Fordow, which involved two ventilation shafts being hit with six GBU-57/Bs each. The bombs were dropped in rapid succession, with each one burrowing deeper and deeper into the facility below. WSMR also happens to contain the site of the first atomic bomb test in 1945.
A map showing the general location of White Sands Missile Range, with an inset giving a better sense of how expansive it is. Holloman Air Force Base, which juts into the southern end of the range complex, is also highlighted. Google Maps
“The plans for the test [at ALT-SHIST] were connected to the Iran war,” CNN said in its story, published yesterday, citing unnamed sources “familiar with the matter.” Beyond that, the planned testing was tied to a “need to develop a way to target Iran’s deepest-buried underground facilities and, likely, simulate U.S. strikes against specific nuclear facilities.”
CNN said it was unable to independently verify whether or not the test in question had occurred.
A formal justification for the sole-source contract to Pate Construction that DTRA released in February did not mention Iran or elaborate on the need for the new testing at ALT-SHIST, but stressed there was an urgent requirement critical to U.S. national security.
“The services required are the immediate and mission-critical enabling services for a ‘real-world’ national security ordinance [sic; ordnance] test. This is not routine construction; it is the highly specialized remediation and operational preparation of a unique, subterranean test facility at White Sands Missile Range (WSMR),” according to the justification document, which DTRA officials had approved on February 24. “The work is a critical path dependency for a live-fire test designed to validate classified capabilities against rapidly emerging threats.”
“As the incumbent contractor for over a decade, Pate Construction Co., Inc. possesses unparalleled and non-transferable tacit knowledge of this specific subterranean test facility. This is a unique Government structure with complex geological and security considerations, and Pate Construction Co., Inc.’s direct, hands-on experience is a tangible asset that no other source possesses,” the document adds. “Pate Construction Co., Inc. is the only contractor that currently maintains a fully mobilized presence at WSMR with the specific heavy equipment and security-cleared operators required.”
CNN‘s piece says apparent work at ALT-SHIST was visible in satellite imagery of the site between mid-February and mid-March, but the outlet said it was unable to confirm whether or not this was related to the contract with Pate Construction.
We reviewed satellite imagery showing work at a remote DTRA/STRATCOM test site that coincided with this contracting action, but we couldn’t confirm its purpose.
The Pentagon didn’t respond to our questions regarding the test, and we couldn’t independently confirm it occurred. pic.twitter.com/k8zyvqWy56
ALT-SHIST is an active test site, and any work observed there earlier this year could well have been linked to other projects. For instance, additional publicly available DTRA contracting documents offer details about planned work on a specific part of the ALT-SHIST site, called Tunnel 7, tracing back to last year, as part of a project nicknamed Millstone Willemite 6 (MW-6).
“This task requires the construction contractor to repair and address the damage to Tunnel 7 caused by prior incidents, in accordance with government-provided designs, specifications, and other contract documents,” according to one contracting document dated September 15, 2025. “The contractor will also be responsible for providing support during testing activities, as well as safing after each test and general support between tests. All tunnel construction must be completed by 15 June 2027.”
“The contractor shall repair a D-shape tunnel and overburden using the D-shape of 20 feet depth and width, inside the tunnel. The contractor shall remove and repair any damage to the ribs, crown, and invert to Tunnel 7,” that same document continues. “After the repair of Tunnel 7, the contractor shall provide general construction support to instrument the tunnel in preparation for testing and then safe the tunnel for re-entry post-testing.”
The schedule for the MW-6-related work indicates that it is separate from what Pate Construction was hired to do at the site back in February.
A view out into a mountainous part of the White Sands Missile Range complex. USACE
ALT-SHIST, which was first established in 1995, is also not the only site DTRA makes use of within WSMR. As its name indicates, it is an alternate to the original SHIST site, use of which began in 1993.
“There was a small addition made to the SHIST site, but that area has been heavily used since 2007,” according to an environmental impact study the U.S. Army Corps of Engineers (USACE) completed in 2015. “The granite bedrock at these sites has been fractured and damaged by extensive testing. There are no existing granite areas adjacent to those sites where similar testing can be performed.”
A map showing the relative locations of SHIST and ALT-SHIST, as well as other test sites, at the northern end of the WSMR complex. USACE
USACE conducted that study in support of the establishment of a new test site at WSMR, simply called the Granite Test Site, to succeed the original SHIST. The Granite Test Site is now in use. There are many more sites within the range complex designed to simulate underground and otherwise hardened targets. This is also just one aspect of the wide array of testing that occurs on a daily basis at WSMR.
All three also appear to have been involved in the testing and development of the Massive Ordnance Penetrator (MOP). 5/10 pic.twitter.com/3aWSYXIZln
However, the presence of large areas of undisturbed granite is a key factor here, as that is particularly relevant for testing weapons designed to penetrate into extremely deeply buried facilities. CNN‘s piece yesterday noted that Iran’s Pickaxe Mountain site is built underneath a granite mountain.
“The mountain in White Sands, they specifically picked mountainous areas that were solid rock, and they used state-of-the-art tunneling technology to make the kind of cavities in the mountain that you would put this kind of [Iranian nuclear] facility in,” Gary Stradling, a retired scientist who previously worked for the Department of Energy’s Los Alamos National Laboratory (LANL), told TWZ in an interview last year. “I think that there is a lot of work that DOD has done in these kinds of facilities that is highly classified, and I couldn’t talk to you about it. This is very sophisticated work. It is not casual, and it’s not speculative. These are real experts who are doing the work. Careful, technical, quantitative work to be able to deliver this kind of war-fighting capability to the services.”
TWZ spoke to Stradling, who did work tied to earth-penetrating munitions at various points during his lengthy career, in the wake of Operation Midnight Hammer.
Me, standing at the edge of a MOP crater on a mountain top at White Sands Test Range, where detailed studies of massive earth penetrators were studied for effectiveness against deeply buried targets by the Defense Threat Reduction Agency (DTRA) —formerly Defense Nuclear Agency.… pic.twitter.com/JYq1V2xO6q
The U.S. military notably did not strike Pickaxe Mountain, which is situated immediately to the southwest of Iran’s nuclear facility at Natanz, during Operation Midnight Hammer. The site also went untouched by Israel during last year’s 12 Day War. American and Israeli forces did not target it during major combat operations earlier this year, either. All of this underscores the targeting challenge the site presents, even for strikes using MOPs, which TWZ explored in detail earlier this year and that has continued to make it a point of particular discussion.
CNN‘s report yesterday was primarily focused on a new assessment of ongoing construction at Pickaxe Mountain from the Center for Strategic and International Studies (CSIS) think tank in Washington, D.C.
Satellite “images reveal more road activity at Pickaxe Mountain in 2026 than at any point in the site’s history,” according to a report CSIS published yesterday. “There is a clear surge in construction activity in 2026 that corresponds to the raising and hardening of the portal accesses, paving of the internal road network, reinforcement around access areas, and the flattening and removal of excavation spoil.”
“This assessment can neither refute nor substantiate the Israeli intelligence claim that Iran moved centrifuges into Pickaxe Mountain,” CSIS’s report also noted. “The continuous construction, the absence of an increased internal security posture, and the lack of observed nonconstruction vehicles and equipment suggest the underground facility is not yet capable of enrichment work, if that is the intended purpose. Further analysis of daily activity would be needed to establish the site’s pattern of life beyond its current construction footprint.”
This is in line with TWZ‘s own previous reporting on Pickaxe Mountain back in July, when there were already signs of growing activity following the establishment of a shaky ceasefire between the United States and Iran the preceding month. As we wrote at that time:
“As with the strikes on Fordow, multiple bombs could be dropped on the same impact point to try to penetrate further down into Pickaxe Mountain. Whether there might be similar ventilation shafts or other weak points that could be exploited at Pickaxe Mountain is unknown. …”
A graphic with details about the employment of MOPs on strikes at Fordow during Operation Midnight Hammer. DOD
“A strike ‘right in the front door’ on the tunnel entrances at Pickaxe Mountain would be another option to at least hamper access to the facility inside for a time. This could be combined with strikes aimed at sections of the tunnel networks that run deeper inside the facility, but that are still close enough to the surface to be reachable by available conventional munitions. For instance, MOPs may not be able to hit the main cavern areas, but they could possibly hit certain tunnels that might lead to them.”
“Even a combined strike on entrances and shallow tunnels at Pickaxe Mountain could still present certain challenges. Firm intelligence on exactly where the tunnels snake under the mountain could be limited, making it difficult to select the best impact points. This is exactly why the U.S. military has been investing heavily in advanced, void-sensing fuzes for MOP and other bunker buster bombs for decades now. These fuzes detect when a munition has penetrated into a sufficiently large space, such as a room or tunnel in an underground facility, to help maximize damage. Fuzes that can effectively ‘count’ floors also help to ensure the munition has burrowed sufficiently deep before detonating.”
…
“…the U.S. military could conduct follow-on strikes on the entry points at Pickaxe Mountain to disrupt any attempt to regain access, as well.”
“Destroying Pickaxe Mountain will be a scientific and targeting challenge,” CSIS also noted in its report yesterday. “It is an underground fortress buttressed by granite. It is one of the only facilities in Iran’s nuclear program that the United States has not yet struck. Structurally degrading Pickaxe Mountain requires more than just kinetic force – it takes actionable high-confidence intelligence and sophisticated mathematical modeling.”
With all this in mind, it is not at all hard to imagine how new testing at ALT-SHIST, as well as other sites, would be essential for exploring additional options for striking Pickaxe Mountain and/or other deeply-buried targets in Iran using MOPs and/or other munitions.
A US Air Force B-2 bomber drops a GBU-57/B MOP during a test. USAF
It’s also worth noting here that GBU-57/Bs are understood to be in relatively short supply, though it is unknown how many of them are currently in inventory. Furthermore, the U.S. Air Force’s B-2 Spirit bombers are currently the only aircraft certified to employ the huge MOPs operationally. There was a report in 2024 that prime contractor Boeing was working to expand its capacity to produce these bombs. A MOP successor, the GBU-76/B Next Generation Penetrator (NGP), is in development now, but prototyping work is expected to run through the end of Fiscal Year 2028.
President Donald Trump certainly has raised the prospect of a strike on Pickaxe Mountain publicly on several occasions since July. He has also repeatedly said that the site is being persistently surveilled, particularly from space.
“There’s another place called Pickaxe Mountain, another nice place. We may have to give them a shot there, too, because someone said there was a little movement at the site,” Trump said yesterday during a speech at a convention the Republican Party is hosting in Dallas, Texas, ahead of this year’s midterm elections. “We don’t need much. We see everything, you know, because of Space Force.”
“We may hit Pickaxe [Mountain] very soon,” the President also said while speaking to reporters at the White House last Friday.
Trump on Pickaxe Mountain: We may have to give them a shot there too. We see everything. We can the see tag on their coat. “Muhammad Al-Fayed” It’s never Muhammad Jones. pic.twitter.com/UQm939OruL
Whether growing activity at Pickaxe Mountain finally prompts a U.S. strike at the site, either alone or as part of a larger operation, remains to be seen. If and when such a mission is launched, past and future testing at sites like ALT-SHIST in New Mexico will all but certainly have played an important part.
Recently released pictures have highlighted a particularly interesting feature on the U.S. Navy’s Arleigh Burke class destroyer USS Mustin: a piggy-backed point defense system with a .50 caliber machine gun mounted on top of a 30mm chain gun. The Mk 38 Mod 4 Gun Weapon System (GWS) is interesting to examine as it offers significant boosts in firepower and other capabilities, including against drones, compared to preceding Mk 38 types.
Yesterday, Destroyer Squadron 15’s (DESRON 15) public affairs office released several pictures showing USS Mustin replenishing at sea in the Indian Ocean, including one highlighting the ship’s Mk 38 Mod 4 GWS (also sometimes confusingly referred to as a Machine Gun System, or MGS). DESRON 15 had put out another group of pictures late last month showing the weapon system being fired during routine training. Mustin was the first vessel in the Navy to receive Mod 4, and still looks to be the service’s only warship with this new version.
USS Mustin fires its Mk 38 Mod 4 GWS during live-fire training in the Indian Ocean in August 2026. USNAnother view of the Mk 38 Mod 4 GWS being fired. USN
“The Mk 38 Mod 4 Machine Gun System (MGS) is an upgrade … to address deficiencies in meeting Counter Unmanned Aerial Systems (C-UAS) requirements,” the Navy has said in the past. “In addition, the upgrade will also improve performance against surface targets and maintain Anti-Terrorist / Force Protection capabilities.”
To take a step back quickly, the original Mk 38 Mod 0 was developed internally by the Navy in the late 1970s and was simply a manually operated deck-mounted 25mm M242 Bushmaster chain gun with a maximum rate of fire of 180 rounds per minute. The M242 was entering U.S. Army service at around the same time as the main armament for the Bradley Fighting Vehicle. A product improved Mod 1 version was the first Mk 38 to actually enter serial production and began to be fitted to Navy ships in the late 1980s.
A Mk 38 Mod 1 weapon system, seen here mounted on the deck of the Oliver Hazard Perry class frigate USS Ingraham in 2003. USN
Following the deadly terrorist attack on the Arleigh Burke class destroyer USS Cole in the port of Aden in Yemen in 2000, the Navy began to acquire Mk 38 Mod 2s, also known as the Typhoon Weapon System, from United Defense (later BAE Systems).
The Mod 2 “introduced remote-control operation and mount stabilization and allowed the weapon to maintain aim even as the ship pitched and rolled. It also added a new electro-optical/infrared sensor and a laser rangefinder, giving the system reliable day/night target performance,” according to the Navy. “Beyond sensors, Mod 2 brought improved ammunition handling and enhanced interfaces – resulting in faster and more reliable engagement. Testing showed it delivered two to three times the strike accuracy of its predecessor.”
The Mk 38 Mod 2 otherwise featured the same M242 gun as the Mod 0/1 types, feeding from a magazine on the mount that could be loaded with up to 168 rounds. It is worth noting here that this is less than a minute of total firing time at the weapon’s maximum rate of fire. However, it can also be fired in single-shot and three-round burst modes, as well as at a lower automatic rate of fire of 100 rounds per minute.
In the 2010s, the Navy began acquiring the Mk 38 Mod 3 from BAE Systems, which added a smaller 7.62x51mm Mk 52 Mod 0 chain gun feeding from a 750-round magazine to the complete weapon system. The Mk 52 is also a much faster-firing weapon than the M242, with a maximum rate of fire of 570 rounds per minute.
The Mod 3 “also incorporated an improved EO/IR sensor that provides 330-degree surveillance capability and multiple fields of view,” per the Navy. “This sensor is decoupled from the gun, allowing operators to scan independently of the weapon and reducing the chance of alerting an adversary that they’ve been detected.”
Over the years, Mk 38 1/2/3 types have been integrated on a wide array of Navy ships, including Ticonderoga class cruisers, Arleigh Burke class destroyers, Nimitz and Ford class aircraft carriers, as well as a host of other amphibious warfare ships and auxiliaries.
The complete Mk 38 Mod 4 also comes with a new Mk 48 Mod 2 electro-optical sight system (EOSS), which is mounted separately on the ship, and a Mk 134 Mod 0 operator console for remote operation. Furthermore, the complete system is designed to be fully integrated with the Aegis Combat System on Arleigh Burke class destroyers like USS Mustin. The optical turret appears to replace the older EOSS system mounted atop the bridge that was primarily used to target the ship’s 5-inch gun. It isn’t clear how these new mounts are individually targeted simultaneously as they don’t appear to have their own optics. Radar targeting through Aegis would be a big advancement for at least some targets types, like drones. The older Mk 38s had their own individual FLIR systems. It’s still possible these new turrets could have a basic boresighted electro-optical system as well.
Various views of the main turreted weapons component of the Mk 38 Mod 4 system. MSI Defense SystemsThe Mk 48 Mod 2 electro-optical sight system (EOSS) that is also part of the Mk 38 Mod 4, but is mounted separately on the ship. MSI Defense Systems
In a July 2022 test, a Mk 38 Mod 4 system “successfully identified, tracked and engaged both surface and aerial targets using live ammunition against fixed and moving targets,” according to the Navy. The event “also highlighted the enhanced fire-control chain, the 30 mm gun’s improved performance and its ability to counter new challenges.”
The USS Mustinreceived its Mk 38 Mod 4 in 2024 and, as noted, this destroyer still appears to be the only ship in the Navy with this new weapon system, for reasons that are not entirely clear. The service had signed a contract with MSI Defense for an initial low-rate production batch of 15 systems in 2023. Other versions of the Seahawk DS30 are in use today on several classes of naval vessels globally, including British Type 23 frigates.
“USS Mustin (DDG 89), an Arleigh Burke class Aegis guided missile destroyer, was the first U.S. Navy warship to receive the upgrade. [The Naval Surface Warfare Center] Dahlgren Division is now finalizing the configuration so the system can be introduced to a wider set of platforms as fleet requirements evolve,” according to an official news item from Naval Sea Systems Command (NAVSEA) just this past February.
Another view of the Mk 38 Mod 4 on USS Mustin. USN
The Navy’s 2027 fiscal year budget request, released earlier this year, makes clear the plan is still to integrate the Mk 38 Mod 4 at least on all Flight IIA and Flight III subvariants of the Arleigh Burke destroyer. The Mk 38 Mod 4 is also set to be a feature on future U.S. Coast Guard icebreakers, which will also have combat systems derived from Aegis.
All remaining Mk 38 Mod 2s in Navy service are being brought up to the Mod 3 standard, as well.
For the Navy’s Arleigh Burke class destroyers, specifically, the Mk 38 Mod 4 reflects broader efforts to improve and expand the shorter-range defensive capabilities of those ships, especially against growing drone threats. An increasing number of these destroyers are also being equipped with new anti-drone interceptors. In addition, the service is working to replace the 20mm Vulcan cannon-armed Mk 15 Phalanx Close-in Weapon Systems (CIWS) on all subvariants in the class with different launchers for RIM-116 Rolling Airframe Missiles (RAM) to further bolster defenses against drones, as well as incoming missiles and other aerial threats.
Mustin may not be the only Navy warship equipped with the Mk 38 Mod 4 for much longer, if the system has not already started finding its way at least onto more Arleigh Burke class destroyers.
Login
