People put away silicon wafers that were on display during an Intel event called AI Everywhere in New York, Thursday, Dec. 14, 2023. (AP Photo)
Canada’s national security agencies are steadily integrating artificial intelligence into their operations, even as oversight bodies examine how these powerful tools are governed and used.
The National Security and Intelligence Review Agency has launched a review of AI use across federal security institutions, formally notifying key ministers and departments. The move reflects growing interest—and concern—around how automation and machine learning are shaping intelligence, policing and risk assessment.
Below is a snapshot of how major federal organizations are deploying AI to support national security, while attempting to balance innovation with legal and ethical safeguards.
Intelligence Agencies Test AI for Analysis Support
At the Canadian Security Intelligence Service, artificial intelligence is being explored as a way to reduce analysts’ workloads rather than replace human judgment.
CSIS plans to launch pilot projects this year using AI for audio transcription, language translation and document review. The agency is also testing a chatbot-style system to assist with drafting, editing and summarizing reports.
Officials stress that all AI-generated outputs remain subject to human verification. Proposed tools undergo algorithmic impact assessments and legal reviews in consultation with the Justice Department before deployment.
Policing Uses AI With Caution
The Royal Canadian Mounted Police has already incorporated limited AI-driven technologies into investigations, including face-matching tools used to analyze lawfully obtained images and videos.
While the RCMP acknowledges AI’s potential to improve efficiency, it has flagged risks around privacy, accuracy and unauthorized access. To address these concerns, the force has created a dedicated AI policy team tasked with building a framework for responsible adoption and internal governance.
Cyber Defence Relies on Machine Learning
Canada’s cyber intelligence agency, the Communications Security Establishment, has been using AI for years to protect federal and critical infrastructure networks.
Machine learning helps identify patterns in massive data sets, detect sophisticated malware and spot cyber threats that traditional antivirus software may miss. According to the agency, AI will become increasingly central to its ability to counter foreign cyber adversaries.
Diplomacy and Intelligence Analysis Go Digital
At Global Affairs Canada, a tool known as “Document Cracker” is helping officials sift through large volumes of diplomatic material.
The AI-powered system allows users to track emerging issues, monitor references to key figures and analyze shifting international positions, enabling faster policy responses in a rapidly changing global environment.
Identity and Travel Security
The Immigration, Refugees and Citizenship Canada uses facial recognition technology within the passport program to verify identities, detect fraud and prevent documents from being issued to ineligible applicants.
Meanwhile, Transport Canada is developing an AI-assisted risk evaluation tool to better assess threats to civilian aviation. The system automates parts of the traditionally manual process of monitoring open-source intelligence and conflict-zone developments.
Oversight Remains in Focus
As AI becomes more embedded in national security work, federal watchdogs and experts emphasize the need for transparency, human oversight and clear rules. The current review by Canada’s spy watchdog is expected to shape how these technologies are governed as their use expands across government.
For now, officials say AI is being positioned as a support tool—enhancing speed and precision—rather than a replacement for human decision-making in matters of national security.
