Canada's Ai Strategy Compounds The Taxation Mistakes The Federal Government Is Already Making

My essay writing assignments in grade school were sometimes handed back with poor grades. “You missed the core ideas, Kim. You need to explore the topic from all important angles. Try again.” My teachers were right when I was honest with myself.

Reading Canada’s new artificial intelligence strategy brought back those memories. My teachers’ voices were loud and clear.

Merriam-Webster defines strategy as “a careful plan or method for achieving a particular goal usually over a long period of time.” Canada absolutely needs a strategy to win the AI race. Investment in AI can create jobs, drive efficiency and productivity, spawn new technologies and assist overall economic growth.

The downsides — labour market disruption, safety concerns, over-reliance by youth — are real, too. A genuine strategy would seriously grapple with both sides.

Unfortunately, all we got was 50 pages of fluff. Instead of focusing on important economic matters such as taxes, the document is devoted to protecting Canadians from AI, Indigenous data governance (whatever that is), diversity, equity and inclusion in AI development and online safety for children.

Some of these may be worthy concerns, but a government that leads with control rather than competitiveness has already revealed its priorities.

Glaringly missing is how the founders, engineers and investors who will build Canada’s AI future will be encouraged to take the necessary steps and risks.

Tax policy is one of the most powerful levers any government uses to shape economic behaviour. It drives investment decisions, determines where talent chooses to live and build and signals to everyone whether a country is serious about competing.

The strategy doesn’t contain any corporate tax commentary — not even patent box proposals despite two years of consultations and a Liberal 2025 election platform promise — personal tax proposals or capital gains ideas.

The sole tax gesture in the strategy document is that by “Budget 2026, the Department of Finance will work with experts to explore mechanisms that encourage Canadians to reinvest gains earned from successful tech companies into new Canadian AI startups .”

Who will those experts be? Why only Canadian AI startups? AI impacts all businesses, not just startups.

Look at what Canada’s competitors are doing. The United States in July 2025 permanently restored full expensing of domestic R&D, reinstated 100 per cent bonus depreciation on qualifying capital investment including AI infrastructure and expanded the qualified small business stock exclusion to US$15 million. The effective tax cost of AI investment dramatically dropped.

The United Kingdom applies a 10 per cent corporate tax rate on profits from patented intellectual property through its Patent Box regime, compared to a standard rate of 25 per cent. Thirteen European Union members have similar regimes. And Estonia — ranked first on the Tax Foundation’s International Tax Competitiveness Index for 12 consecutive years — taxes corporate income only when distributed, not when reinvested.

All those ideas are worthy of consideration.

Instead, Canada’s competitive position on tax is poor. Our combined federal-provincial corporate rate of 23 per cent to 27 per cent, depending on the province, is broadly comparable to high-tax U.S. states, but meaningfully higher than the 21 per cent available in zero-tax jurisdictions such as Texas and Washington, among the fastest-growing AI employment hubs in North America.

It’s also far above Ireland’s 12.5 per cent, which is why so much tech IP ends up there. A promised Canadian patent box was absent from the last budget , the spring economic update and the AI strategy.

The picture is equally concerning on the personal side.

Canada’s top combined federal-provincial marginal rate reaches 53.53 per cent in Ontario and is similar in many other provinces, with the federal top rate of 33 per cent kicking in at $258,482. The equivalent top rate in the U.S. doesn’t apply until income exceeds US$768,700 for a married couple.

For the talent Canada needs most — people earning $300,000 to $600,000 in a sector where that compensation is routine — our tax system is a billboard that reads, “Leave. Or stay away.”

The founder’s exit math compounds the problem. Canada’s lifetime capital gains exemption (LCGE) for qualifying small business shares stands at almost $1.3 million in 2026, while the U.S. exclusion reaches US$15 million. That 11-fold gap shapes where founders incorporate, where companies scale and whether Canada captures any of the wealth its researchers create.

As a further example, if the government is prepared to make a $10-million capital gains exemption permanent for employee ownership trust transfers — a regime that is structurally ineffective — then the existing exemption is simply not a serious number for founders in the AI era.

Absorbing the fiscal cost of a gesture almost no entrepreneur will ever use while refusing to move the one lever almost every entrepreneur can plan for is not a tax policy judgment, it is a political one.

• We shouldn't hate the world's first trillionaire — we should encourage Canadians to emulate him
• Beware tax marketing, it's never as good as it seems

The overall tax tools Canada needs to assist in achieving the AI race include a lower general corporate rate, which could make the patent box idea moot if it were a meaningful reduction, significant reductions in top personal rates, capital gains deferral opportunities, an LCGE expanded to at least $5 million, ideally matching the US$15-million exemption in the U.S., and overall comprehensive tax review and reform .

Canada’s AI strategy is long on words, safety frameworks and consultation, and so obviously designed to achieve political objectives rather than economic ones.

My grade school teachers had a simple standard: did you explore the topic from all important angles and produce logical comments, alternatives and solutions to the issue at hand? If I had submitted this AI strategy for grading, my teachers would have told me to go back to the definition of strategy and try again rather than produce a political brochure.

They also would have given me a glowing grade of D minus.

Kim Moody, FCPA, FCA, TEP, is the founder of Moodys Tax/Moodys Private Client, a former chair of the Canadian Tax Foundation, former chair of the Society of Estate Practitioners (Canada) and has held many other leadership positions in the Canadian tax community. He can be reached at kgcm@kimgcmoody.com and his LinkedIn profile is https://www.linkedin.com/in/kimgcmoody.

_____________________________________________________________

If you like this story, sign up for the FP Investor Newsletter.

_____________________________________________________________