Military Challenges

CRS Report. Replicator's Hardware Hits Targets. Autonomy Lags 6-12 Months Behind. The Congressional Research Service provides a reality check on DIU's flagship program. Only hundreds of systems fielded versus thousands promised. But the real story is the autonomy gap. The numbers tell two tales. Hardware: 80% on track. Autonomy: 60% at best. $1B+ invested across FY2024-2026 with $500M already obligated. Tranche 1.1 delivered Anduril Roadrunners and GA variants. Tranche 1.2 focused on "lethal effects" software. Why the delay? Electromagnetic interference in contested environments. DoD Directive 3000.09 on autonomous weapons creates legal hurdles. AI reliability in jamming scenarios remains unproven. Full swarming without human oversight pushed to mid-2026. The strategic pivot cuts deep. August 27, Hegseth consolidated Replicator 2 into Task Force 401. F-35 sustainment cut 10% to fund attritables. Legacy programs lose to "disruptive" tech with 18-24 month cycles versus 10+ years. Early wins matter. INDOPACOM exercises show 2x effectiveness against simulated PRC defenses. 40% better penetration rates. Cost reductions hit 50% below legacy UAS. But without autonomous swarming, mass loses its edge. Congress lacks visibility. Organizational changes are undefined. CRS urges $2B+ in FY2026 to hit scale. Oversight hearings likely. Three challenges persist. 1. Scalability for 1,000+ unit coordination. 2. Counter-countermeasures as cheap tech proliferates. 3. Training infrastructure for swarm operators. Does your AI work in heavy jamming? Swarm coordination algorithms battle-tested? Replicator proves hardware is easy. Autonomy is hard.

Soldiers, mud and Snow. Operating in war, in a cold climate is difficult. When the weather sits close to freezing water remains liquid. Every footfall, every tyre tread turns the ground into impassable mud. The mud clings to the feet, legs, hands and equipment. It becomes more and more difficult to maintain a level of operational momentum. When movement is prevented the troops become static, coated in mud with wet feet. Static troops are vulnerable to air and ground attack so movement is severely discouraged especially by day. At these temperatures feet do not freeze they absorb water and destroy the tissue and circulation leading to a condition known as 'immersion foot' or 'trench foot'. When left untreated the feet can eventually develop gangrene. The only solution is a period of dryness. Soldiers often wear previously soaked socks around their necks in an attempt to dry them out but wet boots soon return dry socks to their former state. Where soldiers have prepared underground shelters with smoke restricted fires they can get ahead of the game for a short periods. High leg, waterproof boots are a true bonus. Clean water becomes an issue in mud and as a result so does diarrhoea. Soldiers in trenches are often unable to move to latrines to defecate. Trench life is extremely uncomfortable. When temperatures get well below freezing, mobility is increased by the solid ground and frozen rivers. The troops can kick the snow from their boots and remain dry, if not warm. They can dry clothes by freezing and shaking the ice crystals out of them but feet still need a lot of protection; they can freeze to ice. Soldiers often sleep with their comrades feet tucked under their arm pits to defrost cold toes. Comrades sleep cuddled together to stay warm. The regular rotation of troops out from front lines to a short distance in the rear to dry out or get warm, reduces the casualty numbers from cold injuries significantly. The snow lays a blanket of warmth over shelters protecting them from the cold and from view. This also offers some defence against infra-red observation. The downside to snow is that it marks every movement of men and vehicles marking routes and hiding places to arial reconnaissance. Most soldiers prefer dry cold to wet cold and they prefer movement to sitting still. Right now the soldiers in Ukraine are dealing with wet cold and mud. Charities and support groups should concentrate their efforts on anything that will protect the feet and hands. Boots, socks, snow gaiters, thermals and so on. This will do more to keep soldiers in fighting condition and healthy than anything else at this time of year. Slava Ukraini! Who Dares Shares Robin Horsfall

After decades of increasingly rapid offensive maneuver, the Russo-Ukraine war is characterized by World War I-style attrition. The enemy can now detect the slightest movements and attack without notice, resulting in a battlefront locked into defensive strongholds with “soldiers buried in trenches, where even personnel rotations and medical evacuations have become perilous.” This is the result of three main developments. The first is small tactical drones, which are used to target military forces and equipment across air, land, and sea—and even fight other drones. The second is electronic warfare, which now encompasses tracing, jamming, and even taking over drone signals. It enabled an enemy to target back and eliminate specialized, difficult-to-replace crews. And the third is remote-controlled sensors of varying complexity. Generously deployed in contested but undefended “white spaces,” they create protective buffers preventing the enemy from sneaking through. https://lnkd.in/exVA6D7Z

SWARM WARFARE ARRIVES: TÜRKİYE’S KARGU DRONES SIGNAL A NEW ERA OF AUTONOMOUS STRIKE CAPABILITY by Nikola Vračević Türkiye has taken a decisive step into the future of warfare with the successful live-fire test of its KARGU loitering munition swarm, where 20 drones executed simultaneous strikes on multiple targets under coordinated autonomous control. This demonstration goes far beyond a simple drone deployment—it confirms the transition from single-use unmanned systems to synchronized, AI-driven combat formations. Controlled by a single operator issuing high-level commands, the swarm demonstrated the ability to navigate, distribute targets, and strike with precision, effectively compressing the decision-making cycle and overwhelming potential defenses through simultaneous engagement. In practical terms, this represents a shift from drones as tactical tools to drones as integrated combat systems capable of shaping the battlefield in real time. However, despite its technological significance, the KARGU swarm remains a system awaiting true battlefield validation. The test was conducted in a controlled environment against relatively simple targets, without exposure to layered air defense systems, electronic warfare interference, or dynamic combat conditions. Its strengths are clear: scalability, reduced manpower requirements, resilience through decentralized control, and the ability to saturate and confuse an adversary. Yet, its limitations are equally important. These drones lack inherent protection against air defense measures, including anti-air systems and counter-UAS technologies, making them vulnerable in contested airspace. Additionally, their operational effectiveness depends heavily on communication integrity and environmental conditions, both of which can be disrupted in real conflict scenarios. For military planners, the KARGU swarm presents both an opportunity and a challenge. It offers a cost-effective method for conducting precision strikes and disrupting enemy formations, particularly against soft or lightly defended targets. At the same time, its integration into conventional force structures requires careful consideration—doctrine, coordination with other assets, and countermeasure resilience must all evolve alongside the technology. Until it is tested under real combat pressure, the system remains a promising but unproven capability, one that signals the direction of modern warfare while still awaiting its defining moment on the battlefield.

The real challenge of AI and autonomous systems in defence is not buying them. It is understanding how to use them. This article in Foreign Affairs touches on a problem I’ve been raising repeatedly. Much of the conversation around AI in defence still focuses on the technology itself. Far less attention is given to doctrine, accountability, integration, and operational reality. And that is where the real challenge lies. Deploying AI systems in warfare is not like deploying a new app. Lives depend on the outcome. If we still struggle to make fully autonomous cars work reliably in civilian environments, it should at least make us pause before assuming autonomy will somehow function seamlessly in the chaos of combat. And the real challenge is much bigger than the technology. Autonomous systems and AI-assisted tools have to become part of: * doctrine * command structures * operational procedures * governance and accountability frameworks * training and delegation models * data pipelines and validation processes * integration and security * part of operation and mission Without that, technology does not solve problems. It simply introduces new risks. - Who validates automatically detected targets? - Who is accountable when an AI-supported decision goes wrong? - Who labels and verifies the enormous amounts of data needed to train and improve these systems? - Who ensures the models remain reliable when the adversary adapts? These are not theoretical questions. Modern battlefields already generate thousands of potential targets every day through sensors, drones, and surveillance systems. AI can help process that volume of information but the system still needs clear rules for verification, prioritisation, and decision-making. And that requires something many defence discussions still underestimate: adoption across the entire system. Not just introducing a new tool. But integrating it from the operator level all the way to strategic command. Flying a single drone manually is one thing. Operating AI-assisted swarms or autonomous systems connected to ISR networks, electronic warfare environments, and legacy platforms is something entirely different. It requires new skills, new training approaches, new doctrine, new organisational thinking and understanding risks. Technology alone does not create capability. Integration, governance, and operational understanding do. That work is difficult. But without it, buying AI systems risks producing something far less useful than expected. Or worse. #DefenceInnovation #AI #Autonomy #MilitaryTechnology https://lnkd.in/dQhkQwz7

As a former Armor & Cavalry officer it pains me to read this, but important lesson for modern warfare: Eastern Ukraine CNN — Ukrainian crews working on US-supplied Abrams tanks have told CNN of a series of the weaknesses and flaws with the armored vehicles, calling into doubt their utility on the war’s ever-changing frontlines.   Crews trained in Germany said the vehicles – the US military’s main $10 million battle tank used in Iraq against Saddam Hussein’s forces and insurgents – lacked armor that could stop modern weapons.   “Its armor is not sufficient for this moment,” said one crew member, callsign Joker. “It doesn’t protect the crew. For real, today this is the war of drones. So now, when the tank rolls out, they always try to hit them.”   Much of the Ukrainian frontline is now dominated by the use of self-destructing attack drones, tiny and accurate devices that can swarm infantry and even cause significant damage to tanks. The advent of these so called First-Person Vision (FPV) drones, flown by soldiers wearing gaming goggles, has changed the nature of the war, limiting movement and introducing a new element of vulnerability to armored vehicles.   Ammunition is also a problem, like elsewhere on the Ukrainian frontline. They say they seem to have the wrong type for the fight they are in. “What we have is more for direct tank-to-tank fights, which happens very rarely,” Joker said. “Much more often we work as artillery. You need to take apart a tree-line or a building. We had a case when we fired 17 rounds into a house and it was still standing.”   The Ukrainian crew expressed frustration the tanks were made for a NATO style of warfare, in which air power and artillery prepare the battlefield before tanks and infantry advance. Kyiv has long bemoaned its lack of artillery and air power. #ukraine #warinukraine #greatpowercompetition #usarmy #technology #tanks #slavaukraini https://lnkd.in/dTa9cXHy

Why Drone Swarms Are Harder Than They Look The idea of drone swarms—hundreds of UAVs flying together, overwhelming defenses, and coordinating like flocks of birds—has become almost mythical in discussions about the future battlefield. But when you peel back the layers, the reality is much more complicated. Despite years of experimentation, we don’t see drone swarms widely deployed in Ukraine, Gaza, or other conflicts. Why? 1. The Spectrum Problem Every drone needs a radio connection to send and receive signals. Multiply that by dozens or hundreds of drones, and suddenly the spectrum becomes a battlefield of its own. Channels get overloaded, signals interfere with each other, and the swarm becomes easier to jam or disrupt. You can try to decentralize and give each drone autonomy, but then you need advanced onboard AI, which comes with its own limitations, especially around cost and weight. 2. The AI Challenge For swarms to function independently, drones need vision and, especially decision-making, that can handle real-world complexity. That’s far harder than it looks in lab demos. Camouflage, weather conditions, and unpredictable movement make AI recognition. Sometimes you don't even want to strike a target you correctly identify, but wait. 3. The Cost Trade-off If you load every drone with powerful processors, sensors, and autonomy, you’re no longer building cheap, disposable systems. You’re building miniaturized fighter jets. And in war, cost-effectiveness often matters more than raw capability. This is why many militaries today still favor simpler drones: single-use, manually guided, and cheap enough to deploy in numbers. 4. What “Swarming” Already Looks Like Even though we don’t see fully autonomous, AI-driven swarms yet, swarm behavior already exists in simpler forms: Shahed drones launched in waves, where sheer numbers overwhelm defenses. Decoys, flown alongside higher-value systems to confuse radars and exhaust air defenses. These aren’t “Hollywood swarms,” but they are swarming tactics in practice. 5. The Future Is Integration, Not a Silver Bullet The future of drone warfare won’t be defined by a single breakthrough technology. It will be about integrating multiple systems into coherent tactics: - UAVs working in tandem with UGVs. - Artillery units deploying their own reconnaissance and targeting drones. 6. What This Means for Innovators Think six months ahead: the war will evolve faster than your development cycle. So if you develop for what's needed today, when you deliver, you are too late. You need to see around the cordner and build for that. From my chat with Yuri Kozik, Chief Business Development Officer at Resist Labs. Full episode can be found here: https://lnkd.in/dEayrEdE https://lnkd.in/dp3GXRK6

Been thinking about the PLA’s new drone mothership, Jiu Tian, and I’m skeptical. Like much of China’s aerospace tech, it’s hard to separate reality and fair dust. The idea: an RQ-4-sized drone releases swarms of smaller drones/loitering munitions from side bays (oddly, not the bottom) to strike unspecified targets. But there are issues. Animations show the mothership releasing very small drones from above 50,000 ft—where it would fly slow, large, and vulnerable to SAMs. Recall Iran shot down a U.S. RQ-4 Global Hawk at similar altitude in 2019. Some images show quadcopters, which can’t operate in thin air at high altitudes. So either Jiu Tian must descend—risking air defenses—or gravity-drop drones that activate lower down. Even then, range becomes a problem: can the drones fly long enough to hit their targets? Other reports say it’s launching FPV drones, implying each is remotely piloted—a logistical nightmare. With Jiu Tian expected to travel 4,500+ miles, you’d need satellite relays and a 1:1 drone-operator ratio. Highly inefficient. Still, the core idea makes operational sense: a mothership deploying swarms of cheap, “good enough” drones mirrors lessons from Ukraine, where mass has proven decisive. The U.S. has its own concepts, such as RAPID DRAGON, which drops palletized munitions from airlifters near the battlespace. Motherships help expeditionary forces generate mass, useful for overwhelming air defenses or launching surprise strikes. But airborne motherships are big, vulnerable targets—meaning their drones need range and altitude. That suggests the need to use missiles or loitering munitions as effectors (as opposed to quadcopters) , though even then there are real physics challenges to deploying from high altitudes. But the real tech barrier for effective, air-launched swarms going forward is autonomy. Swarm effectors (loitering munitions, kamikaze drones, whatever) must have at least pre-programmed flight paths. But it would be tough to predictively automate a swarm of hundreds to mass on a particular target. What’s really needed is a true, AI- enabled drone swarm that can make group decisions in real time. It’s coming, but we are not quite there yet.

Manuel López (Firestorm’s head of autonomy) and I have been talking a lot about the comms issues that are going to be needed to solved before autonomous swarms that coordinate between platforms becomes the norm. A number of solutions we are seeing/being pitched to empower inter drone communications and swarm behavior today are using communication bands that are being actively jammed in Ukraine. Comms in a highly contested EMI environment is the new normal and thus a crucial piece of the puzzle to solve before widespread adoption of autonomy makes its way to the battle field. We will need to think differently and solve this in an affordable manner to find success with autonomous swarms. https://lnkd.in/g4ffbgw4

In today’s defense ecosystem, everyone’s talking about loitering munitions, swarm drones, and autonomous platforms. These are the visible tools of modern warfare—fast-moving, high-tech, headline-worthy. But the real enabler? Communication. While the drones fly and systems engage, tactical communications—the ability to transmit and receive secure, uninterrupted data and voice across all domains—is what keeps the mission coherent, the units coordinated, and the commanders informed. From my own experience in the field, I can tell you this: no action starts without a green light, and no green light comes without reliable comms. Let’s break down the real-world challenges: 1. GPS-Denied Environments Near-peer conflicts have made GNSS jamming and spoofing commonplace. Without robust fallback systems, even the best positioning or timing systems are blind. HF solutions—properly engineered—offer a resilient, SATCOM-independent layer that operates across thousands of kilometers, providing reliable time, position, and messaging continuity. 2. Urban and Cluttered Terrain In dense cities or mountainous regions, line-of-sight VHF or SATCOM is degraded. Here, self-healing MANET networks shine—especially those built for mobility, multi-hop, and dynamic topologies. Systems like those integrated by Wavestorm (including Creomagic’s advanced mesh nodes) adapt in real time, maintaining secure connectivity without fixed infrastructure. 3. High Throughput Demands for ISR and Video Today’s commanders demand real-time ISR feeds from unmanned platforms—often over extended distances. Traditional narrowband radios can’t keep up. High-bandwidth MANET radios, capable of pushing HD video with low latency, are becoming essential—not just nice-to-have. 4. Contested Spectrum and EW Threats Jammers and intercept tools are evolving fast. Communications gear must now incorporate frequency agility, cognitive routing, LPI/LPD modes, and encryption—not as upgrades, but as base requirements. 5. Disconnected, Disrupted, Intermittent, and Limited (D-DIL) Conditions Humanitarian missions, SOF teams, Arctic patrols—many operations begin where infrastructure ends. HF, VHF, and MANET each serve a role in these D-DIL scenarios. The trick is not picking one, but integrating all—multi-layered, interoperable comms that adjust to the environment in real time. Wavestorm Technologies specialize in these multi-domain communication layers: -HF radio systems for long-range redundancy -VHF solutions for tactical ground and vehicular mobility -Advanced MANET networks for ISR, C2, and mission-critical data flow *All platforms are MIL-STD-certified, hot-zone validated, and optimized for mission continuity under stress. This is not about radios. It’s about delivering information when it matters most. #TacticalComms #MANET #HF #VHF #MilitaryInnovation #EWResilience #DefenseTech #C2Systems #ISR #WavestormTechnologies Canadian Armed Forces | Forces armées canadiennes US Army