Beyond AIS: Why Raw Tracking Data Isn’t Enough for Naval Procurement Anymore

At a Glance

• Naval procurement conversations are framed around vessel tracking, but tracking and intelligence solve different problems, and raw AIS feeds only ever deliver the first half.
• AIS was built as a collision-avoidance protocol, not an authenticated intelligence feed, with no cryptographic verification that a broadcast position or identity is genuine.
• GPS jamming affected ~171,286 distinct vessels globally in Q2 2026, and 3.35 million false ship-to-ship meetings were recorded since Operation Epic Fury began on February 28, 2026, with injected positioning coordinates making vessels appear to rendezvous when no meeting took place.
• Dark activity surged sevenfold quarter-over-quarter, with 3,137 prolonged dark events recorded in Q2 2026 involving 2,157 unique vessels.
• All-source fusion, which combines AIS with satellite imagery, radio frequency signals, behavioral analytics, and other intelligence sources, is the operational answer to the intelligence gap that raw AIS cannot close.
• Fusion is not a rip-and-replace decision, and the intelligence layer that works best is the one that slots into the systems procurement teams already have in place.

Procurement Is Asking for Tracking When It Needs Intelligence

Naval procurement conversations across 2026 increasingly begin with the same request. The service needs better vessel tracking. The requirement documents specify AIS coverage, refresh rates, and geographic zones. The evaluation criteria focus on data volume, latency, and integration formats.

But tracking is not the problem naval intelligence is being asked to solve.

Tracking tells a navy where a vessel is. Intelligence tells the navy what that means and what to do about it. Raw AIS feeds only ever deliver the first half. When the vessels a navy most needs to understand are the vessels that manipulate, disable, or spoof the tracking signal, the gap between tracking and intelligence is where naval risk lives.

AIS Was Never an Intelligence Feed

The Automatic Identification System was built in the late 1990s as a collision-avoidance protocol. It was never designed as an authenticated intelligence feed.

There is no cryptographic verification in AIS. Any transmitter can broadcast any position and any identity. The protocol has no mechanism to distinguish a real broadcast from a manipulated one.

This was acceptable when AIS was used for its intended purpose. It becomes structurally inadequate when navies, sanctions authorities, marine insurers, and commodity intelligence teams rely on AIS data to make consequential decisions. The signal the world’s maritime intelligence architecture has been built on is a cooperative broadcast that adversaries can manipulate at will.

The Scale of the Problem

Windward’s Q2 2026 Maritime Risk Report tracked GPS jamming affecting 171,286 distinct vessels in Q2 2026. Since Operation Epic Fury began, Windward has detected 3.35 million false ship-to-ship meetings globally, in which injected positioning coordinates made vessels appear to rendezvous when no meeting actually took place. 

On a single day, March 20, 2026, false ship-to-ship meetings peaked at 356,773. Independent maritime cyber tracking by Cydome estimates approximately 1,000 GPS incidents per day affecting 40,000 vessels, aligning with the broader picture Windward’s data captures.

Dark activity has surged in parallel. 

Q2 2026 recorded 3,137 prolonged dark activity events involving 2,157 unique cargo ships and tankers over 10,000 DWT, a sevenfold increase in events over the previous quarter. 

Some of that dark activity has legitimate justifications. A meaningful share does not. In every case, the vessel operating dark is not self-reporting its position for a period long enough to complete a port call, a ship-to-ship transfer, or a sanctioned cargo movement.

Q2 2026 saw 275 internationally trading tankers broadcasting the flag of a fraudulent registry, with 22 distinct fraudulent ship registries now identified. 

Around 90% of vessels using fraudulent registries are Western-sanctioned, meaning the identity manipulation is being used specifically to obscure the operational reality that a navy or enforcement authority would most want to see.

The signal that some navies rely on to monitor adversaries is the very signal that bad actors manipulate.

What Multi-Sensor Fusion Adds

Multi-sensor fusion is the operational answer that has emerged over the past decade, and it is what separates intelligence platforms from tracking providers.

Fusion combines AIS with satellite imagery, both synthetic aperture radar (SAR) and electro-optical (EO), radio frequency (RF) detection, dark-vessel detection, behavioral history built on years of proprietary maritime data, and any additional source the operational environment demands. These sources cross-verify each other. 

A vessel that broadcasts an AIS position while SAR imagery shows it in a different location is a vessel whose AIS signal has been manipulated. A vessel that stops broadcasting AIS but continues to emit radio frequency signals is a vessel operating dark rather than genuinely off-station. A vessel whose behavioral pattern across multiple voyages diverges sharply from comparable vessels of similar type and trade is a vessel worth investigating regardless of what its documents say.

The value of fusion is not that each individual sensor is better. It is that no adversary can manipulate all of the sensors coherently across all layers. 

Windward’s Maritime AI™ Platform is built on this fusion approach, with 15 years of proprietary behavioral data underpinning the analytical models that turn raw multi-sensor input into decision-ready intelligence. A tracking provider hands the navy a feed. An intelligence partner hands the Navy an assessment.

Fusion Has to Fit Into What Navies Already Run

Naval commands do not want to rip out existing systems and replace them with a new stack. Existing sensor investments, existing operational workflows, existing integrations with allied and joint force systems all represent high sunk cost and operational continuity.

The intelligence architecture that actually gets adopted is the one that integrates with what procurement teams already run. Fusion should sit as an intelligence layer above existing sensor investments, not as a replacement for them. 

What fusion adds is the analytical layer that turns the sum of those inputs into an operational picture the sensors on their own cannot produce. The Navy keeps its existing infrastructure. It gains the intelligence output that the existing infrastructure was not designed to deliver.

For naval procurement teams evaluating options, this architectural fit is often the decisive factor. A capability that promises fusion but requires system replacement is a much harder procurement decision than a capability that promises fusion and integrates with the existing sensor investments the service has already made.

Where This Conversation Is Heading

The naval procurement conversation in 2026 is shifting. The requirement documents still often specify tracking. The capabilities being evaluated increasingly need to be intelligence platforms. The gap between what the requirements say and what the operational environment actually demands is widening.

Adversary AIS manipulation is not going to slow down. Dark activity is not going to decrease. Fraudulent identity broadcasts are not going to become less sophisticated. The signal the architecture depends on is going to become less reliable, not more, over the next several procurement cycles.

The intelligence architecture that closes the gap is available. The question for naval procurement teams is whether the requirements being written now specify tracking or specify intelligence.

Written by Maya Romi, Maritime Intelligence Content Specialist, Windward.