Four Existential Risks in Naval Defense Tech

Fortunately, One of Them Is China’s

Mar 07, 2025

The U.S. Navy has had one heck of a year. Key shipbuilding programs are stalled, the Chief of Naval Operations is fired, and the Congress is chaotically pushing complex acquisition reform instead of a budget. Despite it all, officers forge ahead. To evolve with threats, Navy leaders are closely watching developments in defense technology. These lessons drive the Navy’s most important investments across today’s readiness and tomorrow’s capability.

But I’m worried the Navy is learning the wrong lessons. Normally, I use this platform to gripe that the lessons are obvious and we’re simply not implementing fast enough. Today, I’ve identified four assumptions about naval defense technology. Any one of them could be wrong – all of them are baked into conventional military thinking – and each of them is existential to get right.

1. Fight from the MOC – But centralized decision-making is slow and dangerous

2. Jamming is Overhyped – Mesh comms are resilient and jammers are targets

3. Navy loves UxVs, but doesn’t use them – Rhetoric and operations need to start aligning

4. The PLARF is a bluff – Carrier killer missiles are capable, but are they combat ready?

In all of these cases, I’m probably wrong. I tried to find key defense tech assumptions and trends that lay foundations for our expectations of the maritime battlefield. If I can convince you that Blue should re-examine a single one of them - we still have time.

Conventional Assumption: “Fight From the Maritime Operations Center (MOC)” is one of the U.S. Navy’s core lessons from last year’s combat operations

Challenge: Centralizing decision-making at fleet headquarters is very risky

With destroyers and aircraft carriers locked in real combat for the last year in the Middle East, U.S. Navy leadership has decided to centralize command and control of warfighting in maritime operations centers. Dubbed “Fight from the MOC” in the Chief of Naval Operations’ Project 33 goals, this line of effort would move more decision-making up to fleet headquarters, away from ship captains and carrier strike group admirals.

The logic behind MOC-centric warfighting has been explained in public by a heavy use of buzz words. “Decision superiority is going to be predicated on our ability to have the right information at the right time to the right warrior at the right classification level,” said VADM Karl Thomas at a conference in San Diego. I’ve talked to active duty officers frequently on this topic, and have encountered a few more specific lines of thinking.

First, cognitive overload on ship watch floors means more brains are needed to process tactical combat information and turn it into warfighting decisions. Since ships are crammed, and all hands are already working all the time, you can’t add watchstanders on the ship. Putting ashore watchstanders in the loop can relieve the cognitive burden afloat. When I last stood watch managing the top secret battle map on an aircraft carrier, the ashore watchstanders seemed to have plenty of time on their hands.

Second, advanced capabilities used in real combat often require the protection of special access programs. On the aircraft carrier, even with a full intelligence staff and multiple SCIFs (a room where Top Secret information lives), there is barely space for the existing people and decision-making. Highly classified decisions often occur behind curtains, or with the protection of only loud Taylor Swift music, so uncleared sailors nearby cannot hear. On land at sprawling fleet headquarters, hundreds of sailors can split up into special access siloes, each with their own war room and a watch officer. The admiral and senior staff officers can come and go as needed. On land at the MOC, reliable internet and career civilian experts give warfighters access to all of human knowledge. At sea, sailors sleep under a flight deck, and face shipboard stressors from equipment maintenance to sexual harassment.

UUV Command & Control Example:

Take the hypothetical example of controlling a fleet of UUVs (unmanned undersea vehicles) in combat. On an aircraft carrier with only five SCIFs, one of these SCIFs has the right personnel, with the right polygraphs to operate the UUV program. As it turns out, these UUVs have been stealthily loitering at an undersea charging station, near the adversary, and their existence is highly sensitive.

So there is a strict quota on the number of people on the planet who can know they’re there. This means that only five people on the aircraft carrier can know about them. If the admiral is one, and one of his action officers two, that leaves three quota spots left for sailors who can actually stand watch and operate the UUVs. These sailors can get read into this program, and one of the three could always be on watch. But in order to bring up the screen to control the UUVs on the admiral’s behalf, they would have to kick out the other watchstanders from their SCIF, where they sit elbow-to-elbow. Since one of these watchstanders plays a critical air defense role, and another is actively tracking adversary communications, leaving their stations unmonitored is a non-starter.

This scenario assumes that the fleet of UUVs is relatively simple to manage – low cognitive burden – and not that they need undivided attention to manage, which would take the watchstander’s brain away from other critical duties.

Challenges with combat cognitive overload and special access programs at sea – these seem like the primary drivers of Fight from the MOC. Additionally, officers are trained that they may delegate decision-making, but never responsibility. Delegation is hard, especially when stakes are high and decisions must be made as fast as missiles fly.

The problem with Fight From the MOC: it centralizes decision-making.

Tension: The CNO explained to Defense Scoop “the MOC will be the “center” to how the Navy fights in a distributed manner.” But – centralized and distributed are opposites. The Navy has been trying to realize a vision of distributed maritime operations for years, and now is trying to centralize the C2 for this disaggregated concept of operations.

Such MOC-central decision-making has two primary risks.

First, it creates a node in a hub-spoke information flow that the adversary can target for systemic effect – a center of gravity. In the Middle East, the Navy’s MOC in Bahrain is safe from the Houthis. In a WW3 scenario fighting the PLA, no operations centers are out of ballistic missile range. Even if a Chinese adversary doesn’t have the political will to strike Hawaii with an ICBM, a centralized MOC offers a prime target for soft kill. Power outages, cyber attacks, car bombs, and poisoning would all be easy shots at a land-based decision-making hub.

Fighting from a centralized operations center on land, even if it’s a new location, makes aircraft carriers seem safe and agile.

Second, decisions ashore are slow and disconnected from the reality of combat. In Navy Intelligence Officer Basic Course, we were briefed from day one that the military forces of authoritarian regimes have much more centralized decision-making structures than ours. Politicians in democracies trust military officers, who trust their junior officers, because they are a-political professionals in a meritocratic system. High-trust, decentralized organizations make decisions fast. Low trust bureaucracies, where all the information flows back to the center, have many decisions to make, and suffer from the same cognitive overload they probably worried about in the first place. Or worse.

Even if the second point is wrong – a three star with a well-rested watch team probably does make better decisions than a one-star with a tiny, stressed staff – a MOC is a big target. There is a reason Ukrainian combat operations centers inside targeting range are highly decentralized: big targets burn.

However, Navy leadership is pushing forward with Fight from the MOC. The Red Sea lessons are “an incredibly valuable warm-up for us in the high-end fight,” an active duty surface warfare officer told The War Zone.

Conventional Assumption: Jamming can deter, degrade or even defeat drones.

Challenge: Resilient mesh networks of many small platforms are un-jammable.

There has been a lot of emphasis on electronic warfare since Ukraine kicked off. From Kursk to Kyiv, the jamming battle has raged across every frequency.

Overnight, updates can tweak algorithms that automate frequency hopping on software-defined antennas. In weeks, units hop between hardware to reach whole other parts of the spectrum. Sometimes they’re evolving to evade the adversary’s jamming. In other cases, they dance defensively – they know they must be able to jam the next inbound drone. The last one came too close.

So pervasive and powerful is jamming that Russians and Ukrainians have resorted to tethering drones, and controlling them with fiber optic cables. Tethered drones are un-jammable because they do not rely on any wireless signal.

In combat, the U.S. Navy has applied these lessons. They have evolved quickly, bringing the first over the air updates to their weapons systems – updates that perhaps tweaked how the weapons interacted with the electromagnetic spectrum. And last month, the Navy announced that they were splashing Houthi drones without firing a shot, likely via jamming or spoofing.

With lessons from Ukraine and their own battlefield experience against the Houthis, the U.S. Navy’s way ahead seems clear. The fleet must have high-end jamming capability. Russians, Ukrainians, and the well-armed People’s Liberation Army have “formidable” electronic warfare capabilities. So we, too, must be able to jam across the spectrum, around the world, around the clock – just like the old recruiting ads.

Except that’s not how it works.

Two tech trends just as powerful as the barrage jamming in Bakhmut suggest that jamming is not dominant.

First, jamming everything, everywhere, all at once would take many antennas with infinite power. A single incoming missile or drone can certainly be jammed. A few could be fried, spoofed, or deceived. A local area on a battlefield can be barrage jammed. But any semblance of a swarm or a mesh network can’t really be jammed efficiently over the amounts of time and space a Pacific battlespace will entail.

USV Swarm Example:

Take the example of a dozen small USVs patrolling a choke point. The vessels are split into six groups of two. They all have Starlink. They all have a software-defined radio that can jump between different parts of the spectrum, and communicate with any other USVs within a few kilometers. A destroyer facing off with them – assume the destroyer knows where all six pairs are – would have to jam down six different lines of bearing. On each line, the destroyer would have to hit the Starlink frequency and the line of sight frequency more powerfully than any of the friendly signals hitting it. Starlink seems tough to jam. Line of site comms between directional antennas – unless the jammer is physically between them – are notoriously tough to jam. Even if the destroyer is successfully cutting off a few of the USV pairs from the other pairs and from space, the un-jammed USVs will be able to upload and download from the cloud and pass instructions over line of site to the others, when possible.

This is a very simple scenario where each USV only has two types of comms, and there aren’t UUVs, UAVs, missiles, or crewed vessels supporting the USVs. Any force that expects to fight an advanced adversary should assume that the adversary will have a decentralized web – not twelve but hundreds of antennas – linking their forces. They won’t all be jammed.

Second, counter-targeting is more important than jamming. Jamming a large part of the spectrum, with a lot of power, over a lot of time is highly visible. Adversary cryptologists – they have passive sensors from space to ships to cell phones – will copy-paste the jammer’s coordinates into their favorite targeting chatroom and the jamming unit will receive a volley of long range missiles.

In short, jamming is very noisy and offers great targeting data. Jamming is also ineffective against mesh networks. Continued investment in electronic attack is essential, but assuming it is anything more than a pre-requisite to survive on the modern battlefield is folly.

3) Maritime Drones

Conventional Assumption: Unmanned maritime systems are key to defeating Russia in Europe and China in the Pacific.

Challenge: But in today’s combat ops, the Navy doesn’t feel a need to test and learn with unmanned tech

Last week, NATO’s Task Force X issued a solicitation for small USVs, UUVs, and UAS it needs to control the Baltic and the Straits of Gibraltar. Last year, the U.S. Navy stood up Task Force 66, with the goal of using small unmanned vessels to jam up Europe’s constrained bodies of water and pin down the Russian fleet. If scrappy Ukrainians can do it, why can’t U.S. and NATO forces? This demand signal for cheap, asymmetric unmanned capability extends beyond Europe to every theater. The Navy is investing heavily in Robotic and Autonomous Systems, as the DoD is in Replicator, to deter or defeat a Chinese crossing of the Taiwan Strait.

[The Task Force X UxV Solicitation is a great read, check it out here]

Despite all the rhetoric on UxVs, fleet operations suggest UxVs to be either un-important or un-ready. The Navy has spent the last year in combat, against a low-end adversary, but has not used the combat conditions to test these new unmanned vehicles. In the Civil War, the Navy tested prototype ships in combat, knowing they needed to test, learn, and iterate – in real combat.

The same leaders who talk repeatedly about how much they’ve learned from the Red Sea and Black Sea – and rightfully preach “train like we fight” – these leaders are the ones saying unmanned capability is important. How do they know, if they’re not using the capability they’re buying? Rear Admiral Mattis, in one of the most informative discussions I’ve found anywhere, explained at CSIS that USV capability has moved “beyond experimentation” thanks to the “Black Sea Battle Lab.”

As an avowed optimist on unmanned capability, I’ve bet my career on building a USV and getting the Navy to use it. My actions speak for my belief in the capability. But Navy leaders who talk about USVs, UUVs, and UAVs at conferences – until they integrate these capabilities with the manned fleet and test them in combat – bely their own rhetoric with their actions. If American USVs aren’t ready for some Houthis, why should the vastly more capable PLA be deterred by them?

In the drone factory where I worked in Kyiv, soldiers home for the weekend would visit on Sunday before returning to the line. They’d observe a test in the parking lot – if it worked, they’d take the tech with them. The tech would go from demo (~TRL 7) to trenches (~TRL 10) in a week.

Two strong arguments exist for not testing blue UxVs in Middle East combat conditions. One is that doing so would give away knowledge of the capability to China and Russia. The Houthis would share this intelligence. This argument would hold for covert capabilities, but such stealth doesn’t make sense for capability you acknowledge and hype up at conferences. Alternatively, the UxVs are in earlier stages of exercise testing, and are not yet ready for the realistic test: combat operations. This argument is tough to buy into because of the numbers of UxVs the Navy is already buying – numbers that suggest they’ve thoroughly tested the tech and trust it. There is a third argument too – that the Middle East combat conditions are not relevant for the other theaters – but this one doesn’t deserve more typeface.

Conventional Assumption: China’s ballistic missiles are kryptonite to US Navy surface ships

Challenge: What if they’re not

Vowing that they would never again be bullied by U.S. Navy Carrier Strike Groups in the 90s, after one too many strike groups sailed into the Taiwan Strait, the PLA’s missile modernization drive began. Today, the PLA operates an entire service branch, equivalent to the U.S. Air Force or Army, dedicated to rockets and missiles. And within the PLA Rocket Force (PLARF), the crown jewels are the carrier killer missiles DF-21 and DF-26.

With ranges of ~1,800 nautical miles and ~3,600 nautical miles, these missiles mean U.S. surface warships and especially U.S. aircraft carriers cannot operate near mainland China in wartime. With hundreds – open source googling estimates 1000+ DF-21 “carrier killers” and 250+ DF-26 “Guam killers”– of missiles that could fly from anywhere in mainland China, even a carrier strike group’s air defenses could quickly be overwhelmed.

U.S. planners must assume that $2B destroyers and $14B carriers cannot be risked against salvoes of missiles that cost just $10M each.

But what if the PLARF is not as proficient in practice as it is capable on paper?

Untested in combat – an ironic, and unexpected similarity between Chinese ballistic missiles and U.S. maritime drones – the whole PLA surely has much to learn as an institution. The rocket force could conceivably be the cushiest, most corrupt part of the PLA, as recent purges suggest. Last year, Chinese missiles were discovered filled with water, in place of fuel. These rocket forces are likely conducting more and more realistic training, a trend across the PLA. But to my knowledge, they have not hit live targets at sea or faced the decision-making challenges of deception and degraded comms that combat would bring.

What if the PLARF fires several salvoes, over a hundred missiles, in the first day of a war, but hits only pre-positioned decoys left by U.S. forces for such a purpose? What if the NSA has planted tracking devices or deadly code on PLARF launchers? What if American spies have been paying PLARF generals for their rocket fuel the whole time?

If Chinese ballistic missiles don’t work on the first day of a conflict, and then on the second, and then on the third… how much doubt will seep into a brittle system? How much doubt can such a system take?

Russia, with its vast missile arsenal, fired huge salvoes into Ukraine as the invasion commenced. Against dispersed forces, they had little effect. The Russians made many mistakes in February of 2022, but they never assumed long range strike capability would carry them to victory. The PLA was wise to develop capability to neuter the mighty U.S. surface fleet. They’ll know too that conventional long range strike has never proved decisive in a war.