Artificial Intelligence in North America (2025 -2030)

The U.S. decisively led in AI research and innovation, releasing 40 significant new AI models in 2024—outpacing all rivals and cementing its preeminence in both technical leadership and research output. U.S.-based researchers contributed 173 of the world's 100 most-cited AI publications between 2021 and 2023, underlining the country's intellectual dominance across the global AI ecosystem. The U.S. also recorded 1.8% of all new job postings referencing AI ¹

AI-related U.S. investment reached nearly $159 billion in 2025, which comprised 79% of global AI funding.² The San Francisco Bay Area received $122 billion, and leading foundation model companies drew over $80 billion. ³
Venture capital activity was marked by a surge in “megarounds” (over $500 million each), massive M&A deals, and an overall shift toward funding established hyperscalers and top-tier startups.

The public sector played a pivotal role: U.S. federal AI investment in 2023 reached $831 million, and the number of federal agencies with AI-regulatory or policy responsibilities grew from 42 to 59 in 2024, demonstrating increased regulatory attention and institutional adaptation.⁴ States increased AI-focused lawmaking, though federal action is moving toward national preemption of potentially fragmented state policy.⁵

The U.S. ranked first worldwide for government AI readiness, thanks to comprehensive national strategies and public–private collaboration.⁶

Canadian federal and provincial governments committed over C$2 billion (2025–2030) for sovereign AI compute resources, public sector capability, and national data residency—a strategy to ensure Canada’s independent AI infrastructure and data governance).⁷

Industry investment surged with Microsoft announcing a CAD $19 billion investment plan (2023–2027), including $7.5 billionin the subsequent two years targeting new data centers and cloud infrastructure).⁸

Canada launched the "AI Assist" program to enable SME adoption of AI and expanded national initiatives in talent upskilling, AI literacy, and equity.⁹ Policy frameworks such as the federal Directive on Automated Decision-Making and voluntary industry codes articulate a human-centered, ethical foundation for domestic AI adoption.¹⁰

The lapse of the proposed Artificial Intelligence and Data Act (AIDA) resulted in a predominance of voluntary codes and stronger provincial-level laws (Quebec, Ontario), as well as federal consultations exploring privacy and human rights alignment.¹¹

By 2025, 78% of North American enterprises had adopted AI in at least one key business function, a dramatic leap from 55%in 2023. The share of firms using generative AI (genAI) doubled to 71% in the same period).¹²

The technology sector showed the highest AI adoption rate at 77%, followed by Media & Entertainment at 69%, and Financial Services at 63%. In Healthcare, nearly all Chief Information Officers indicated plans to implement generative AI by 2026. The Mining and Agriculture sectors also demonstrated significant and widespread uptake of AI technologies.¹³

In the retail sector, more than 95% of all customer service interactions were AI-powered. The retail industry also witnessed a 39% compound annual growth rate (CAGR) in AI-driven CRM and dynamic pricing.¹⁴

Across all industries (including mining, agriculture, and the public sector) integration of AI tools and automation is now widespread.¹⁵

North American data center vacancy reached an historic low of 1.6% in H1 2025, driven by hyperscaler and generative AI demand. Demand is stressing power, site, and permitting systems across the U.S. and Canada.¹⁶

Data centers now account for over 4% of U.S. electricity consumption, with projections suggesting this could rise to 12% by 2028.¹⁷ Increased private and public sector capital is being deployed to expand both data center capacity and grid transmission, with hundreds of billions in capital earmarked for infrastructure .¹⁸

U.S. AI-related job postings increased by 68% year-on-year in late 2024,¹⁹ comprising 49,577 unique roles in Q4 alone. All sectors reported higher AI exposure, impacting professional requirements across the U.S. and Canada.²⁰

84% of U.S. LinkedIn members hold jobs where at least 25% of tasks could be automated by generative AI. Up to 10% of all jobs are deemed "high risk" of replacement by 2029, though millions of new roles are expected to emerge, demanding large-scale upskilling and retraining.

Only 24% of Canadian workers have received AI-specific training, below the global average of 39%, driving the launch of new federal education and literacy initiatives, and major investment in national research institutes. Both Canada and the U.S. are emphasizing inclusive, human-centric adaptation strategies.²¹

Jobs exposed to AI frequently see wage premiums(especially in high-skill fields such as energy, healthcare, and professional services)with higher earnings and new opportunities in these domains.²²

The U.S. federal government is asserting primacy in AI regulation, issuing executive orders that preempt state laws and fast-track permitting and federal legal frameworks. While states have enacted over 100 AI-specific measures in health, employment, and consumer protection, these are increasingly subject to federal override.²³

In Canada, with the failure to pass AIDA, the governance landscape is led by voluntary codes, provincial laws (such as Quebec Law 25 and Ontario Bill 194), and ongoing consultations towards national privacy and human rights safeguards.²⁴

Both countries are moving toward transparency, continuous risk assessment, safety, and human-centric guidelines as pillars of AI regulation.²⁵

AI-exposed jobs are seeing notable wage premiums, as AI elevates value creation, particularly in energy, healthcare, and professional services sectors.

All surveyed sectors (mining, agriculture, services, manufacturing, government) reported increased AI adoption in 2025. The pace of adoption is reshaping public and private expectations regarding automation, trust, and oversight.²⁶

Policy and regulatory debates are becoming central, as society grapples with balancing innovation, safety, global competitiveness, and sustainability.²⁷

By 2026, leading AI labs (including OpenAI) intend to introduce models functioning as "AI research interns," capable of meaningful collaborative and autonomous research. By 2028, roadmaps foresee AI "autonomous researchers" able to drive scientific and technical breakthroughs with minimal human supervision, albeit with peer review and validation pending.²⁸

A surge in sector/domain-specific AI models (“verticalization”), the rise of neuromorphic and edge computing, and expanded focus on fields such as healthcare, automotive, and robotics will widen use cases and drive new adoption.²⁹ Enterprises are transitioning from pilot projects to full-scale deployment.³⁰

The capital required to support future AI expansion (data centers, chips, power grid) is forecast to reach hundreds of billions, possibly trillions, between 2026–2030. There is a material risk that infrastructure investments could outpace real productivity gains, raising concerns about an "AI bubble" if adoption falters or ROI is not sustained.³¹

Energy use by data centers could climb to 12% of U.S. electricity by 2028, exerting significant pressure on existing grids, regional permitting, and sustainability strategies. This will demand regulatory innovation and new approaches to siting, grid expansion, and environmental standards.³²

Routine knowledge work automation will intensify, driving gains in productivity and enabling new classes of AI-driven employment, but also increasing polarization between displaced jobs and those commanding wage premiums. Economic benefit will depend on the speed and scale of workforce adaptation and upskilling.³³

Wage growth in AI-exposed sectors will likely continue, but job displacement risks will persist absent significant investment in human capital .

Both the U.S. and Canada are expected to develop more granular, risk and safety-based governance, potentially drawing on evolving debates in the EU and international arenas.³⁴

Technical oversight, mandatory incident reporting, transparency standards, and continuous compliance will become standard regulatory practice. Ongoing monitoring, assessment, and agile regulatory bodies will be critical for resilience.³⁵

Growth scenarios range from robust productivity and ROI (best case) to supply–demand mismatches and infrastructure overbuild (worst case). Outcomes depend heavily on continued technology adoption, regulatory agility, workforce adaptation, and capital market discipline.

Bold timelines for fully autonomous AI researchers are as-yet unproven outside company projections, with broader validation and oversight required.