Federal government spent more than $800M on AI contracts, licensing over 3 years

Ottawa has spent more than $800 million on AI-related technology since 2023, including a $350 million Dayforce contract to replace the Phoenix pay system and a $240 million investment in Cohere. Public Services and Procurement Canada and Innovation Canada were the top spenders, followed by National Defence at $83.7 million and the Canada Revenue Agency at $29.9 million. The data highlights broad federal adoption of AI, but the article is primarily a spending disclosure rather than a market-moving policy shift.

Analysis

The main market takeaway is not that governments are “using AI,” but that public-sector procurement is becoming a demand backstop for enterprise software vendors at a time when private enterprise budgets are still being scrutinized. That matters because government buyers tend to be stickier, slower to rip out installed systems, and less price-sensitive once a workflow is embedded, which can extend revenue visibility for the winners even if headline AI enthusiasm cools. The larger second-order effect is that large incumbents with compliance-ready products can outcompete smaller pure-play AI firms that lack procurement-ready security, auditability, and data residency features.

DAY is the cleanest near-term beneficiary, but the more durable implication is across the broader “AI middleware” stack: identity, workflow automation, HR, finance, and security vendors that can package AI as a governance upgrade rather than a speculative model layer. The article also signals a multi-year budget reallocation inside the public sector, where AI spend is likely to grow fastest in defense, tax, and operations rather than consumer-facing use cases. That should support vendors with regulated-market exposure and depress the odds that the market’s most hyped model companies capture all of the economic value.

The contrarian risk is that this is a procurement spike, not a smooth runway: governments often front-load pilots and then delay scale-up once implementation, union, and privacy issues surface. If oversight tightens, some contracts could be re-scoped into lower-margin services rather than recurring software revenue, which would compress the multiple investors are willing to pay for the “AI adoption” story. The right horizon is months for sentiment, but years for budget persistence; near term, expect volatility around any audit, implementation delay, or cost-overrun headlines.

From a positioning perspective, the opportunity is less about chasing the broad AI basket and more about owning regulated workflow winners while fading names that rely on generic AI adoption without contractual lock-in. The market likely underestimates how much of public-sector AI spend will flow to incumbents with distribution and compliance rather than to standalone AI labs. That creates a good setup for relative-value trades rather than outright beta longs.

Key Decisions for Investors

• Long DAY on pullbacks over the next 1-3 weeks; treat this as a catalyst-supported re-rating trade if government implementation milestones stay on track. Risk: any delay in the public-service rollout could cap upside and compress the AI premium.
• Pair trade: long regulated workflow/software names vs. short speculative AI infrastructure exposure over 3-6 months, because procurement favors auditability and integration over model novelty. Use this to express the view that enterprise capture will accrue to incumbents, not pure-play AI narratives.
• Sell out-of-the-money calls on DAY into strength for 1-2 month horizons if implied volatility inflates on additional procurement headlines. This monetizes the likely mismatch between long-dated policy adoption and short-dated enthusiasm.
• Avoid chasing broad AI basket momentum for 3-6 months; focus on companies with government/compliance distribution advantages. The risk/reward is better in names with contracted revenue than in names dependent on discretionary budget expansion.
• Set a tactical alert for any audit or privacy controversy in federal AI spending; that is the highest-probability catalyst to reverse the current sentiment and hit multiples across regulated AI beneficiaries.