For a long time, artificial intelligence in the military sphere was portrayed as a collection of experimental projects, often operating in isolation and deployed in specific contexts. In recent months, this pattern has changed radically. According to reports by Haaretz and confirmed by the Israel Defense Forces themselves, the AI infrastructure developed during operations in the Gaza Strip has not been decommissioned, but transformed into a stable platform, now also operational in other theatres such as Iran and Lebanon.
The point of discontinuity lies not in individual algorithms, but in the overall architecture. The IDF (Israel Defense Forces) has completed a transition to a centralised model based on a military cloud, designed to collect, standardise and make large quantities of operational data immediately usable.
Overcoming information fragmentation
In the past, analytical capabilities were distributed across different units, each with its own tools and information flows. This approach, common to many complex organisations, inevitably led to bottlenecks: duplicate data, systems that did not communicate with one another, and long delays in transforming raw information into a coherent operational picture.
The new model, however, aims to unify heterogeneous sources, from sensors to intercepted communications, from video streams to electronic signals, correlating everything in real time. The result is a shared view of the operational context, continuously updated and accessible to the various units without the friction typical of the past.
Time as a competitive advantage
This change has a direct impact on the most critical factor in military operations: time. The platform enables a drastic reduction in the interval between data collection, analysis and decision-making. Activities such as planning complex operations, prioritising targets or tracking missiles and drones take place continuously and with a level of real-time updating that was previously difficult to achieve.
It is not a matter of delegating the decision to the machine, but of providing decision-makers with coherent and contextualised information within timeframes compatible with high-speed operational scenarios.
A modular and integrated ecosystem
From a technical perspective, the infrastructure is not a monolithic system. It consists of several specialised modules that work in an integrated manner. MapIt provides a dynamic three-dimensional map of the battlefield, whilst FLOW enables the rapid creation of operational dashboards via a no-code approach. These are complemented by agent-based AI systems, designed to break down complex tasks and collect data autonomously.
Development is coordinated by the Matzpen unit, which reports to the ‘Bina’ AI division, established in 2025. The choice to use fine-tuned open-source models suggests a strategy focused on operational flexibility and direct control of the technology.
AI and offensive intelligence
At the same time, artificial intelligence has also become central to offensive intelligence activities. According to the Washington Post, specialised units such as Unit 8200 use classified platforms capable of analysing targets’ digital footprints and tracking their movements in real time.
This approach allows for more effective synchronisation of multi-domain operations and ever-closer integration with autonomous systems, such as drones equipped with computer vision. Compared to previous systems, the leap forward lies not so much in the use of AI for targeting, but in the level of integration and scalability of the entire ecosystem.
A data-driven transformation
From a strategic perspective, what is emerging is a structural transformation in the way operations are conducted. Human control remains central, but AI amplifies analytical capability, increases operational scale and drastically reduces coordination times.
The competitive advantage is not linked solely to military power in the traditional sense, but to the ability to process large volumes of information with speed and precision.
The implications for cybersecurity
This evolution does not concern the military domain alone. The dynamics observed are similar to those emerging in the world of cybersecurity. Data centralisation, for example, is already a fundamental requirement for managing complex and distributed IT environments. Without a unified view, identifying an incident becomes a fragmented and ineffective process.
The same applies to advanced correlation. The exponential increase in security events makes manual analysis increasingly unsustainable. AI-based systems enable the identification of patterns and anomalies that would otherwise remain invisible.
Another key element is the reduction in decision-making time. In the cyber domain, this means detecting and containing an attack before it spreads, thereby limiting its operational impact. Consequently, increasingly sophisticated forms of automation are emerging, in which certain phases of analysis and response are carried out in an assisted or semi-autonomous manner, whilst still maintaining human control over critical decisions.
In conclusion
The artificial intelligence infrastructure developed by the IDF represents a significant case study, not so much for the geopolitical context, but for the operational model it introduces. It is a concrete example of how data management is becoming the true enabling factor, both in modern conflicts and in digital security.
In both cases, artificial intelligence does not replace human expertise, but amplifies its capabilities, making a level of complexity manageable that until a few years ago would have been simply unmanageable.
Analysis by Vasily Kononov – Threat Intelligence Lead, CYBEROO
