While the world watches Microsoft and Anthropic, Canada is having its AI moment

In a 2026 report by Thomson Reuters, the organization asked more than 1,500 professionals whether their organizations use AI. 40% answered yes, nearly double the 22% from a year earlier. 

The same survey found that only 18% of organizations track the return on that spending.

That gap between use and proof is where tech investors should be looking. Most of them are looking somewhere else. 

Here are three signals investors should be reading. 

The giants get the headlines

For much of the past year, the AI story has been told through two companies: Microsoft (MSFT) and Anthropic. Microsoft committed to invest up to $5 billion in Anthropic in November of this past year, and Anthropic in turn committed to buy $30 billion of Azure computing capacity.

In July, Microsoft recorded a $3.2 billion gain on that stake, a number that ran on business pages around the world. Claude models now sit inside Microsoft 365 Copilot. Anthropic shared this week that companies including Gorilla Logic had landed select partner status in its Claude network.

Today, Microsoft announced that Sonata Software had earned the 2026–2027 Microsoft AI Business Solutions Inner Circle award, marking the company’s sixth Inner Circle recognition. “The ability to embrace AI to deliver value and enterprise velocity will differentiate organizations in an AI led era,” said Chief Revenue Officer Anthony Lange in a company statement.

Every model release, funding round and earnings call between the two has become shorthand for where AI is heading.

The deals are often struck in Redmond and San Francisco. However, whether they pay off is decided in places like Toronto and Montreal, where customers either put the tools to work or don't.

Microsoft, for one, knows this. It said last December it would invest more than $7.5 billion in Canada over two years, part of an estimated $19 billion between 2023 and 2027, with new capacity starting to come online in the second half of 2026. That plan included keeping Microsoft 365 Copilot processing and sovereign controls on Canadian soil, and a partnership to offer models from Toronto's Cohere on Azure. 

A look into Canada as an AI center

Canada, in other words, is not just a sideshow to the Microsoft and Anthropic story; it’s one of the places where that story gets tested. It also has a government trying to win the test on purpose. After the April election last year, Prime Minister Mark Carney's government named Evan Solomon as Canada's first Minister of Artificial Intelligence and Digital Innovation. 

This June, Ottawa released its refreshed national strategy, AI for All, which is expected to shape Canada's approach to AI policy at home and abroad. The Prime Minister's Office describes it as a five-year plan built around trust, opportunity and Canadian sovereignty, with targets for growth, AI-related jobs and adoption. 

The private money is moving too. Tech investments alone drew just over $2.4 billion USD in seed to growth funding, with companies such as Toronto-based Provision playing an important role here. 

What comes next will depend on which tools survive contact with demanding buyers. This week offered a Canadian case study. Altorney, a Baltimore-based legaltech company, announced that McCarthy Tétrault, one of Canada's leading business law firms, has deployed its MARC document intelligence platform. The details of that deal line up neatly with two of Ottawa's three priorities, trust and sovereignty. The third, adoption, is playing out among the firms hired to put AI models to work.

Trust and sovereignty

Document review is the most expensive grind in litigation. Teams of lawyers read thousands of documents to decide what is relevant, what is privileged and what contains personal information. 

MARC is built differently, in that for every document it classifies, it tells them not just what matters, but why. Susan Wortzman, a partner at McCarthy Tétrault and president of MT>3, the firm's e-discovery division, singled this out. "MARC provides detailed rationale for every coding decision, making it easy for lawyers to validate its accuracy and spend their time applying complex legal judgment," she said in the announcement. 

According to Shimmy Messing, Altorney's CEO, "Our goal is not to replace legal judgment; it is to give exceptional legal teams a faster, more explainable foundation for applying it."

The second detail in the company's announcement matters as well. "Throughout, all data remains secure within our own private environment, giving our clients confidence that their data is securely managed." That flexibility answers the question that stalls most enterprise AI pilots. Who holds the data, and in which country? 

Gartner said in January that it expects 35% of countries to be locked into region-specific AI platforms by 2027, up from 5% today. Canada is among those moving fastest. Its Sovereign AI Compute Strategy, backed by $2 billion over five years in Budget 2024, aims to reduce reliance on foreign cloud infrastructure and keep Canadian data in Canada. 

For tech investors, this is a trend we should expect to see more of.

AI adoption as a signal to decision makers

In addition to the above,  a model on its own does very little inside a company. Someone has to help businesses unlock data value, accelerate digital transformation, and achieve sustainable impact. Executives such as André Gagné are leading the charge here. 

In Canada, AI for All takes direct aim at business adoption, especially among small and medium-sized companies, and the money is starting to flow: PrairiesCan announced more than $10.2 million in July for six Manitoba organizations. Smaller firms rarely have engineers to spare, so most of that adoption will be bought from outside.

This is the less-covered half of Anthropic's year. The Microsoft deal made headlines; the work of getting Claude into ordinary companies did not. In March, Anthropic launched its Claude Partner Network with $100 million USD committed to training, technical support and joint market development for the firms that do implementation work. The network has expanded to firms such as Gorilla Logic.

"Anyone can build an agent," said Drew Naukam, Gorilla Logic's CEO. “Which makes engineering standards matter more than ever.”

The requirement for production deployments isn't only development that investors should take note of. For example within the talent lifecycle space, Kirsten Moorefield of Cloverleaf Labs recently noted in a webinar on AI, “How do we better support people who want development but are too busy for it? How do we better solve some of these age-old issues that we're all facing.”

That gap is a services and solutions market. Across the globe, it comes with a policy tailwind, and the firms leveraging AI are positioned to win it.

What investors should ask 

The Microsoft and Anthropic headlines tell investors who is building the models and who is paying for the compute. They say much less about who will make money putting AI to work. For that, look at the questions a Canadian buyer now asks, which map closely onto Ottawa's own priorities. Can the product explain its output to the person who has to sign off on it? Can it run where the customer's data already sits, ideally on Canadian soil?

McCarthy Tétrault's answer on all three was yes. Canada's strategy is betting that more of the country's companies can say yes too.