WHITEPAPERS

India’s Maritime Blind Spot: Why 2.4 Million Square Kilometers of EEZ Cannot Run on AIS Alone

What’s inside?

    The Geometry of the Problem

    India’s maritime security challenge is, first and foremost, a geometry problem.

    Seven thousand five hundred kilometers of coastline. A 2.4 million square kilometer Exclusive Economic Zone. The Nine-Degree and Six-Degree Channels threading through the Lakshadweep and Maldives archipelagos. The Andaman Sea sitting astride the northern approaches to the Strait of Malacca. The Palk Strait and Gulf of Mannar to the south. And to the west, the Arabian Sea — connecting the Gulf of Aden, the Strait of Hormuz, and the Makran coast to India’s western seaboard.

    That western corridor is where the picture has become most contested. In Q1 2026, Operation Epic Fury and the subsequent closure of the Strait of Hormuz by IRGC forces drove transit volumes down 97% — from an average of 120 daily transits to just 3.6. Crude and condensate loadings in the Gulf collapsed from 16.6 million to 4.3 million barrels per day.

    Strait of Hormuz transits began to flatline on Mar 2, 2026. Source: Windward Maritime AI™ Platform.
    Strait of Hormuz transits began to flatline on Mar 2, 2026.
    Source: Windward Maritime AI™ Platform

    A Voluntary System in an Involuntary Threat Environment

    AIS became mandatory for large commercial vessels under SOLAS in 2002. Its purpose was collision avoidance, not intelligence. It was designed on the assumption that vessels would transmit honestly, continuously, and in good faith.

    That assumption holds for bulk commercial traffic operating normally. It does not hold for the operating environments that define India’s maritime security challenge.

    The structural problem is regulatory scope. AIS is mandatory only for commercial vessels above 300 gross tons on international voyages. The vessels posing the highest risk in the IOR — dhows, fishing boats, small cargo traders, coastal speedboats — fall entirely outside that requirement. Many carry no AIS at all. Others operate on Class B transponders that transmit intermittently and can be legally switched off. In the Arabian Sea narcotics corridor, along the Makran coast approaches, and across the Andaman Sea, the vessels moving the most sensitive cargo are precisely the ones that AIS-based monitoring is least equipped to see.

    65–75% of Gulf crude exports go to Asia. India sits in the middle of that supply chain, and the disruption arrived without warning in the monitoring systems of organizations that depend on AIS to tell them what is moving and where.

    Across the same geography, dark fleet activity continued to expand. Windward’s tracked dark fleet reached 2,108 vessels in Q1 2026, up 45 from Q4 2025, with 48% sanctioned and 1.77 million barrels per day of Iranian crude flowing through it daily. That fleet operates across the same sea lanes that connect the Gulf to Indian refineries and Indian Ocean ports. The vessels moving that crude have a demonstrated operational interest in not being accurately tracked.

    SAR imagery showing two tankers with AIS switched off loading at Kharg Island terminals on April 26, 2026, at 02:37 UTC. Source: Windward Remote Sensing Intelligence
    SAR imagery showing two tankers with AIS switched off loading at Kharg Island terminals on April 26, 2026, at 02:37 UTC.
    Source: Windward Remote Sensing Intelligence

    Where the Signal Gets Manipulated

    In Q1 2026, as conflict escalated in the Middle East, 978,000 GPS jamming events were recorded globally in a single quarter — 98% of them concentrated in the Gulf.

    AIS positions falsely placing vessels on Iranian land due to GPS jamming and signal distortion detected across the Gulf region. March 2026.
Source: Windward Maritime AI™ Platform
    AIS positions falsely placing vessels on Iranian land due to GPS jamming and signal distortion detected across the Gulf region. March 2026.
    Source: Windward Maritime AI™ Platform

    More than 1,100 vessels were affected. The practical result: AIS effectively became unreliable as an intelligence source across the region that generates the majority of IOR energy traffic. Vessels were broadcasting positions their own navigation systems could not verify. Monitoring systems built around that broadcast were producing a picture with no reliable ground truth beneath it.

    Beyond jamming, identity manipulation is endemic among the dark fleet vessels transiting toward and through Indian Ocean waters. In Q1 2026, 290 tankers were broadcasting fraudulent registry flags — operating under 20 fraudulent registries, up from 18 in the prior quarter. Eighty-eight percent of ships using fraudulent registries are sanctioned. These are not edge cases. They are the operating profile of a fleet designed to be invisible to systems that take what vessels broadcast at face value.

    Ship-to-ship transfers, a well-documented tactic for obscuring cargo origin, leave traces in AIS data only if both vessels are transmitting. When one or both go dark for the transfer, the gap is treated by broadcast-dependent systems as a coverage anomaly. In the Arabian Sea and western IOR, it is frequently evidence of a deliberate operation.

    EO image shows two ship-to-ship meetings. One (above) between two transmitting vessels, and another (below), semi-dark meeting between a Tanzanian-flagged service vessel and a larger dark vessel, with its AIS turned off.
    EO image shows two ship-to-ship meetings. One (above) between two transmitting vessels, and another (below), semi-dark meeting between a Tanzanian-flagged service vessel and a larger dark vessel, with its AIS turned off.
    Source: Windward Maritime AI™ Platform

    India established the Information Fusion Centre for the Indian Ocean Region at Gurugram in 2018 with a clear mandate: serve as the region’s primary node for maritime domain awareness, fusing information from partner nations, agencies, and white shipping data into a common operating picture.

    The ambition is exactly right. The challenge is that the information environment the IFC-IOR was built to work with has become significantly more contested. AIS manipulation, GPS jamming, identity fraud, and dark activity have all increased in sophistication and frequency in the approaches feeding IOR traffic. The cooperative signals framework that white shipping agreements depend on is under sustained pressure from actors who understand which signals to manipulate and when.

    Fulfilling the IFC-IOR’s mandate, and India’s broader SAGAR commitment to being the region’s preferred security partner, requires an intelligence architecture that does not depend on the honesty of the vessels being monitored. That means access to data that cannot be simultaneously manipulated: drawn from independent sensors, across independent technical domains, operated by independent providers. 

    When 978,000 jamming events compromise AIS in a single quarter, the agencies that maintain an accurate picture of IOR traffic are the ones whose architecture does not treat AIS as the terminal data source.

    The Sources That See What AIS Cannot

    Multi-Source Intelligence is not a single system. It is an architecture: the deliberate integration of every observable data stream about vessel presence and activity into a unified picture that behavioral models can then interpret for operational meaning.

    Satellite and SAR imagery provides direct, independent evidence of vessel presence regardless of what a vessel is broadcasting. A dhow conducting a transfer in the Arabian Sea with its transponder off is still physically present and observable. SAR operates through cloud cover, day and night — removing the monsoon season gap that degrades optical collection across the northern Indian Ocean for months at a time.

    A flotilla of Chinese military vessels captured on SAR imagery, March 18, 2026. AIS showed only 4 vessels transmitting, but the image reveals 5 additional vessels sailing in proximity.
Source: Windward Maritime AI™ Platform
    A flotilla of Chinese military vessels captured on SAR imagery, March 18, 2026. AIS showed only 4 vessels transmitting, but the image reveals 5 additional vessels sailing in proximity.
    Source: Windward Maritime AI™ Platform

    During the Q1 2026 Hormuz closure, EO imagery confirmed vessel positions that AIS had rendered meaningless: Windward’s Remote Sensing Intelligence showed vessels transmitting false locations, with contemporaneous imagery confirming empty waters at the declared coordinates.

    Satellite imagery of three non-AIS transmitting VLCCs at Kharg Island on April 11, 2026. Source: Windward Remote Sensing Intelligence.
    Satellite imagery of three non-AIS transmitting VLCCs at Kharg Island on April 11, 2026.
    Source: Windward Remote Sensing Intelligence

    Radio frequency emissions extend detection to vessels that have suppressed every intentional broadcast. Navigation radar, satellite communications terminals, and crew welfare systems generate RF signals independent of the AIS transponder. RF collection can approximate vessel location even in full AIS darkness — and RF emission fingerprints can link a vessel across voyages even when its name, flag, and MMSI have all changed. For the dark fleet operating in IOR corridors, where identity cycling is standard practice, RF fingerprinting is one of the most reliable tools for maintaining track continuity across a vessel’s operational life.

    Device presence signals — from smartphones and satellite communicators aboard — are independent of vessel systems and largely beyond the control of anyone seeking to suppress a vessel’s electronic profile. In the Riau Archipelago, where Windward tracked 100 non-transmitting vessels in Q1 2026, device signals can confirm activity that no other source is capturing.

    Camera intelligence at chokepoints — the approaches to major Indian ports, the Palk Strait, the Nine-Degree Channel — creates identification opportunities that cannot be avoided without bypassing the location entirely. For vessels that have cycled through flags and MMSI numbers, physical observation at a chokepoint restores ground truth that electronic records have obscured.

    Open source intelligence — registration databases, flag state records, port authority communications, company filings — provides the ownership and commercial context that makes sensor data intelligible. The 290 tankers broadcasting fraudulent flags in Q1 2026 are not identified by their transmissions. They are identified by cross-referencing what those transmissions claim against what independent records, imagery, and behavioral history reveal.

    Entity Resolution: One Vessel, One Record

    The foundational challenge in fusion is entity resolution: determining that multiple observations from different sources refer to the same vessel, and building a single persistent identity record around it.

    A vessel may appear in satellite imagery under one name, broadcast a different MMSI on AIS, be registered to a company operating under a third name, and appear in OSINT records under a fourth. Its RF emission fingerprint may link it to a vessel that officially no longer exists. When 88% of the vessels using fraudulent registries are sanctioned, and those vessels are actively cycling through flags across fraudulent registries spanning Netherlands Antilles, Guyana, Guinea, and Madagascar — as Windward’s Q1 2026 data confirms — the entity behind the signal is exactly what is being hidden.

    Entity resolution is what makes manipulation visible. A vessel appearing in two locations simultaneously is not a data anomaly. An MMSI shared between two distinct hulls is not a quality issue. A vessel whose RF fingerprint matches a scrapped hull is not a database error. Each is an intelligence lead that a broadcast-dependent system would file as noise.

    From there, behavioral assessment becomes possible. The patterns that matter in the IOR are well understood: loitering in the western Arabian Sea without a declared port call; dark periods followed by positional discontinuity inconsistent with the elapsed time; repeated proximate meetings between the same pair of vessels at sea; draft changes consistent with cargo transfer where no port call occurred. None of these findings emerge from position data alone. They emerge from position data fused across sources, assigned to a persistent identity, and run through behavioral models that know what normal looks like — and recognize when it is not.

    Entity Resolution: One Vessel, One Record

    The Operational Standard India’s Maritime Domain Demands

    India’s strategic posture in the IOR is forward-leaning by design. SAGAR, the IFC-IOR, and the Indian Navy’s expanding blue-water presence all reflect a judgment that India’s security interests extend well beyond its territorial waters, across the full arc of the Indian Ocean, from the Gulf approaches in the west to the Andaman Sea in the east.

    That posture requires intelligence infrastructure that matches its geographic ambition. In Q1 2026, 13 vessels were boarded and detained globally — a 160% increase over Q4 2025 — with all 13 previously flagged by Windward for risk linked to Russia, Iran, Venezuela, or Syria, and 85% determined to be falsely flagged. The enforcement environment is accelerating. The agencies that act first are the ones whose picture does not depend on what the vessel chose to broadcast.

    A monitoring architecture built around a single cooperative data source inherits every vulnerability of that source. The structural answer is not better AIS coverage. It is fusion across independent sensors, resolved against a persistent vessel identity, placed on a unified timeline, and interpreted for behavioral meaning.

    Multi-Source Intelligence is the architecture that closes that gap. Not by replacing AIS, but by ensuring that when AIS is manipulated, degraded, or simply absent, independent sources continue to produce an operational picture. Every observable vessel, regardless of what it chooses to broadcast. Every behavioral deviation, regardless of whether the vessel intended to reveal it. Every identity manipulation, surfaced by comparing signals that were never designed to be compared.


    See Multi-Source Intelligence in Action