Canada’s new National Security Strategy, the first in more than twenty years, places strategic autonomy and national resilience at the centre of the country’s approach to security. Canada must now decide which defence capabilities to acquire and how those capabilities will adapt once fielded. In electronic warfare (EW) and counter-uncrewed aerial systems (counter-UAS), that means putting artificial intelligence at the core of the sensing and decision architecture rather than adding it later as a back-end analytical tool.
Military operations have relied on the electromagnetic spectrum since WWII, but today’s dependence is far greater. Drones, radars, tactical communications, satellite links, and smart munitions all rely on spectrum, meaning modern forces must navigate a congested, contested spectrum against constantly adapting adversaries.
Russia’s war against Ukraine provides the clearest illustration. Both sides continually develop electronic warfare techniques to disrupt opposing systems while adapting their own to evade interference. Frequencies, waveforms and software can change within days or weeks. The electromagnetic spectrum has become a decisive arena of modern warfare, and operational advantage increasingly depends on interpreting new activity and adapting faster than an adversary.
Unlike traditional hardware built to catch known signals, an AI-native architecture treats the spectrum as a continuous data environment. It integrates radio-frequency (RF) and sensor data to spot patterns and flag unfamiliar threats, while leaving final decisions to human operators.
Counter-UAS operations illustrate this shift clearly. Inexpensive, adaptable drones now operate with minimal radio emissions or standard control links, rendering static signal matching obsolete. Defence now requires a multi-sensor intelligence layer that tracks behavioural changes and helps operators assess evolving intent in real time.
Our experience at Zighra illustrates the opportunity. IntentHunter, our AI-enabled decision-support platform for counter-UAS operations, combines RF, drone and other sensor observations to identify abnormal behaviour, assess potential intent and provide explainable information to operators. It is designed to complement existing sensors and support human judgment, helping operators interpret changing threats more quickly.
While Canada is experimenting with faster counter-UAS procurement through the Defence Drone Initiative and IDEaS Sandboxes, its main electronic warfare efforts remain tied to slow hardware-focused capital acquisition cycles. This legacy model cannot keep pace with continuously evolving AI software. To stay ahead, Canada should consider making three practical changes. First, create a continuous acquisition pathway for AI and electronic warfare software. Second, grant domestic industry recurring access to operational data and test ranges to validate sovereign systems and third, structure procurements around modular architectures that safeguard Canadian control over critical software, data, and intellectual property.
There is also the question of sovereignty. Canada’s strategic autonomy depends not just on where hardware is built, but on retaining operational control over the underlying technology. Owning platforms means little if updating software, integrating new capabilities or adapting to emerging threats requires foreign permission.
This does not mean Canada must develop every component domestically or turn away from trusted allies. Canada already has considerable expertise in AI, radio frequency technologies, sensors and autonomous systems. The opportunity is to build a sovereign AI-native layer that can work across sensors and platforms while integrating with allied systems. Sovereignty in this context means retaining the ability to understand, modify, update and sustain the critical intelligence layer without depending entirely on a foreign supplier.
Putting AI at the core is not a substitute for capable hardware. It is how Canada can keep that hardware relevant as threats change. The National Security Strategy provides the direction; Budget 2026 can build the foundation and provide the procurement and sustainment models to treat operational AI as a living capability. If Canada succeeds, it can move beyond acquiring EW platforms and build electronic defence that learns, adapts and improves at the speed of the threat.
Deepak Dutt is the founder and CEO of Zighra, an Ottawa-based Canadian defence AI company developing explainable AI for electronic warfare, counter-UAS operations and spectrum awareness.