Canadian Business Leaders Urge Ottawa to Move Fast on Balanced AI Rules, Boost Incentives, and Build a Sovereign Cloud

Canadian business leaders want swift, balanced AI regulation - here's what government can do now

Canadian business leaders are asking Ottawa to implement a federal AI regulatory framework as soon as possible, according to the KPMG in Canada AI Business Adoption Survey. Ninety-two percent support a framework that is adaptive and light, given the speed of AI advances. Ninety-four percent say policy must be reasonable, uncomplicated, and enforceable - and aligned with international standards to speed adoption.

What leaders are asking for

• Act now on a federal AI framework (92 percent support).
• Keep the approach adaptive and light to keep pace with technology (92 percent).
• Make rules reasonable, uncomplicated, and enforceable (94 percent).
• Align with international standards to streamline adoption (94 percent).
• Stand up cross-jurisdiction AI sandboxes with allies like the UK and EU (88 percent).

Leaders expect a nuanced, balanced approach that moves quickly. The core trade-off: protecting people and markets, minimizing complexity, and still incentivizing AI development in Canada.

A practical regulatory runway

One path is to move fast with a phased approach. Start with lighter rules that lean on existing privacy, human rights, IP, and consumer protection standards. Adjust over time for enforcement and clarity, and learn from jurisdictions like the EU and UK.

• Reference points:
  • UK: AI regulation - a pro-innovation approach
  • EU: European approach to AI (AI Act overview)

Incentives: necessary, but more is needed

Eighty-seven percent say the new incentives for AI adoption - including a 20 percent tax credit for SMEs - were necessary and overdue. Still, 91 percent want stronger tools: low-interest loans, grants, and targeted R&D funding to drive real deployment.

Leaders also want to keep Canadian talent and IP at home. They see stronger incentives as essential to staying competitive in a global AI economy.

Digital infrastructure and sovereign compute

Ninety-two percent want the government to prioritize AI compute: data centers, cloud capacity, and advanced chips. While $2.5 billion is earmarked over the next two fiscal years, 88 percent say it falls short of what's needed.

There's strong support for building a sovereign cloud to reduce exposure to cross-border legal and geopolitical risks. Strong encryption and key management help, but departments will still need to weigh trade-offs across cost, performance, and residual risk.

What federal teams can do this quarter

• Publish interim guidance: use existing privacy, human rights, IP, and consumer protection laws as the baseline for AI use and enforcement.
• Define risk tiers and impact assessment templates for AI systems used in government and contractors.
• Stand up cross-border AI sandboxes with the UK and EU to pilot test standards, audits, and disclosures.
• Set clear liability and documentation rules for high-impact uses (e.g., eligibility, safety, hiring, compliance).
• Map compute needs across departments; coordinate procurement with shared services for GPUs, secure cloud, and data pipelines.
• Accelerate sovereign cloud planning: governance model, funding, site selection, and security controls.
• Tune incentives: pair the SME tax credit with low-interest loans, matching grants, and targeted R&D to de-risk deployments.
• Define success metrics: time-to-compliance for pilots, model audit turnaround time, cost per compliant deployment, and incident rates.

Policy design notes for a balanced approach

• Keep it simple: short, enforceable rules first; detailed standards via guidance and sandboxes later.
• Be interoperable: align documentation and risk categories with EU/UK to cut red tape for exporters.
• Protect outcomes, not methods: set outcome-based obligations so innovation isn't boxed in.
• Enforce with evidence: focus audits on data provenance, testing coverage, and model change logs.
• Reward good actors: fast-track approvals for organizations with strong governance and audit trails.

Survey scope

The KPMG in Canada AI Business Adoption Survey captured feedback from 750 business leaders between August 15 and September 3. The message is consistent: move quickly, keep the rules clear, and build the infrastructure to compete.

Resources

• EU AI Act overview
• UK AI regulation white paper
• Complete AI Training: practical upskilling paths by job function