The Communications Security Establishment turns 80.
Or, its busy
The Communications Security Establishment, Canada’s signals intelligence agency, with a combined foreign intelligence and cyber security mission, turns 80 this year. It’s an under-stated theme in the just-released CSE annual report for 2025-2026. [1]
CSE is twice as old as CSIS and probably less than half as well understood by the Canadian public.
For an excellent short history of CSE, you can turn to Bill Robinson’s account, available here:
https://opencanada.org/marking-70-years-eavesdropping-canada/
For those of you who don’t know Bill’s work, check out his unique blog on all things CSE, at “Lux Ex Umbra,”
https://luxexumbra.blogspot.com
Canada’s SIGINT organisation began with a mere 62 employees drawn from the remaining, dwindling ranks of a wartime traffic analysis centre called the Joint Discrimination Unit (JDU) and the civilians employed on code-breaking at the Examination Unit (XU). They showed up for duty on September 3, 1946, in their first home on the third floor of the LaSalle Academy off Sussex Drive in downtown Ottawa.
That this small band (of brothers, almost entirely) would have work to go to at all was a close-run thing, as officials in the government debated whether or not Canada needed a peacetime signals intelligence capability. The alternative, seriously considered, was a return to the pre-war practice of relying on the British for such intelligence.
Once launched, Canadian signals intelligence never looked back, though there were some difficult moments during the Cold War, including a long climb to achieve any real intercept capability against Soviet communications targets from the Canadian Arctic, the early loss of access to high-level Soviet codes, and the momentous adjustment to the coming of the computer age (in the course of which CSE acquired its first super-computer). Through it all, Canadian SIGINT was the country’s main contributor to intelligence sharing among what came to be called the Five Eyes partnership (with the UK, US, Australia and New Zealand). Canada was critical to its expansion beyond the early UK-US agreement in 1946, called BRUSA. The on-going value of the FVEYs, whatever the difficulties Trump administration actions have placed in its path, is extolled throughout the report.
CSE grew rapidly in the post 9/11 period as it came to take on a significant counter-terrorism mission. It continues to expand, to meet what the report calls a “complex” threat environment, including that posed by organized crime. It benefitted from new, and substantial, infusions of cash in Budget 2025. The annual report notes its current establishment of 4,178 people represents an 8 percent increase from the previous year. No CER (Comprehensive Expenditure Review) for it.
CSE now bills itself as “Canada’s Digital Front Line of Defence.” There is a lot happening on that front line, whether generated by foreign states, their proxies, non-state actors, or organised crime. As the report nicely puts it “these threats do not sleep, nor do they retreat in bad weather, nor do they take breaks on statutory holidays.” (p. 4)
The cyber vulnerabilities—the so-called “attack surface”-- are immense.
You can turn to the CSE annual report to get a sense of the current scale of CSE activities, neatly summarized in a page of stats (p. 6). To summarize the summary, here are some key figures:
Number of foreign signals intelligence reports = 3,976
Number of cyber security incidents responded to = 3,216
(The cyber security incidents are evenly divided between those targeting Government of Canada systems and those directed at other entities.)
Supply-chain risk assessments = 1,772
As I read the 55-page report, three things, in particular, caught my attention.
One is a greater willingness to provide some insight into the kinds of “active” cyber operations CSE is conducting under its new 2019 Act. These are operations designed to pre-empt and disrupt a foreign cyber threat at its source. The report briefly describes two such operations: one against an overseas criminal organization involved in brokering fentanyl precursors (p. 14); the other against a violent extremist group (p. 15)
A second theme reflects how CSE is actively moving into the “national” security space. More reporting is being done on Arctic issues. CSE is helping the three northern governments (the Northwest Territories, Yukon and Nunavut) to protect their systems from hacks and cyber espionage. It notes how the CSE’s Cyber Centre (Canadian Centre for Cyber Security) began to deploy its sensors on the information networks of these governments starting in 2024-2025 to “detect malicious cyber activity in devices at the network perimeter and in the cloud.” (p. 21). The report also notes efforts to draw the federal government closer together with the provinces and territories to advance an understanding of cyber threats and create a stronger coordination of efforts across the country (p. 22). This initiative includes, we are told, “building information-sharing channels”--never easy when it comes to top secret information. Whether this effort to enhance partnerships will extend to CSE assisting the Alberta government in fending off digital foreign interference threats in the course of the October referendum on secession is unknown. Maybe we will learn more in next year’s annual report (?).
A third theme concerns efforts to respond to the changing technological environment, including on-the-horizon quantum computing, and now omnipresent AI. There is a capsule account, valuable as an overview, of how CSE is exploring and using AI tools to improve its signals intelligence analysis (see p. 29). The report cites one example of an AI-enabled tool, the equivalent of a CSE Swiss army knife. This tool is under development, the report says, to allow CSE to translate data from over 200 languages, filter data from large datasets, use chat-based interactions for “conversational queries,” (sounds better than the stuff we consumers have to suffer through) and apply a broader ”semantic” analysis of data beyond the traditional key word searches, including visual topic modelling.
But for all its high-tech tools, the CSE annual report continues to insist on the importance of its human workforce. As the CSE Chief, Caroline Xavier, tells it, “Over those 80 years [of CSE history], it has always been our people who have made the difference” (p. 5). There is still a person (or now a few thousand) at the heart of the person-machine interface. Long may it last.
[1] Communications Security Establishment, Annual Report 2025-26, released June 29, 2026, https://www.cse-cst.gc.ca/sites/default/files/cse-annualreport-2025-2026-e.pdf


A very insightful overview. What caught my attention is the "active" disruption of foreign cyber threats. This suggests that Canada has been moving more and more to offensive capacity.
Linguist-analysts were a very important part of CSE from the earliest days. The focus at the time was on hiring personnel who had a solid understanding of Russian and other Soviet bloc languages, and training those who did not have those skills.
CSE's needs shifted dramatically as the Cold War wound down in the late 1980s, when the organization began to hire dozens of linguist-analysts who had superior skills in various other languages.
As Wesley Wark noted, CSE experienced a surge of employees after the 9/11 attacks on Washington DC and New York City. Managers at that time determined that, as in the other Five Eyes countries, there was a much greater need for analysts who could understand the communications habits of identified hostile actors (especially designated terrorist organizations). This entailed a greater need to understand telecommunications patterns.
However, the need for these important skills was sometimes conflated with a push to retrain linguist-analysts which, according to some observers, may subsequently have undermined the organization's capacity to retain people who comprehend foreign languages and understand the foreign cultural and political contexts that affect the actions of targeted entities.
Today, CSE, like other elements of the Government of Canada's intelligence community, arguably needs to devote greater attention to the cultivation and retention of staff who possess the in-depth linguistic and cultural skills required for properly interpreting foreign intelligence collected by the organization.