Banking on AI: Generative AI Adoption in Canada’s Financial Sector

The Issue

Canada’s financial sector is one of the country’s largest and most economically significant industries, and among the most exposed to AI. Many core occupational tasks in the sector align closely with the capabilities of large language models.

At the same time, financial institutions operate within strict regulatory, privacy, and risk-management frameworks. They must balance productivity gains with obligations around security, explainability, and public trust.

This report provides evidence on workforce exposure, real-world usage, and task suitability to inform the responsible adoption of generative AI across the sector.

What We Investigated

Banking on AI examines how generative AI may reshape work in Canada’s financial sector and how institutions are responding.

The project conducted:

• A novel task-level occupational exposure analysis to identify which financial roles are most aligned with generative AI capabilities.
• An analysis of real-world usage data from Anthropic’s Claude and Microsoft’s Copilot to understand how generative AI is actually being deployed.

Together, this evidence supports sector-specific recommendations on where generative AI can add value, where it introduces operational or compliance risk, and how organizations can implement the technology responsibly within existing regulatory frameworks.

What We’re Learning

High sector-wide generative AI expose. The occupational exposure analysis shows that 98 per cent of financial sector workers are in jobs highly exposed to AI technologies, a much larger share than the overall Canadian workforce. Among these workers, about 24 per cent are in roles where AI is more likely to assist tasks (high complementarity), while a larger share (73 per cent) are in occupations where AI has a higher likelihood of replacing specific tasks. The risk of task replacement is concentrated in business, finance and administrative roles as well as sales and service positions.

Generative AI is not suitable for core numeracy task. Analysis of real usage data from Anthropic’s Claude and Microsoft’s Copilot shows that generative AI is not particularly effective for core numeracy or high-precision financial functions, largely due to the inherent inaccuracy in generative AI systems. In contrast to traditional AI systems that excel at numerical work, generative models struggle with accuracy in these areas.

Use cases: customer service and ancillary business support. Generative AI is being deployed the most (both in actual usage patterns and in terms of suitability) in tasks associated with front-line customer interaction (such as answering customer questions) and ancillary business support tasks that are less dependent on numerical precision.

Why It Matters

The scale of AI exposure in Canada’s financial sector means that generative AI will influence how work is organized across one of the country’s most economically significant industries.

The report finds that generative AI is not effective for core numeracy and high-precision financial functions. Instead, its current strengths lie in tasks like front-line customer interaction and ancillary business support. These findings clarify where the technology can realistically be deployed today, and where caution is still warranted.

For financial sector leaders, this report identifies specific, evidence-backed use cases for generative A deployment. For policymakers and regulators, understanding these boundaries is essential in a highly regulated environment where privacy, security, and trust are paramount.

In a sector that underpins much of Canada’s economy, insights into the practical limits and opportunities of generative AI are critical for informed decision-making.

State of Skills:
Unleashing AI into the Skills Development Ecosystem

FSC-supported AI tools have bolstered outcomes in skills matching, career development guidance, and recruitment. The overall effectiveness of these tools was underpinned by recognizing and mitigating the inherent bias and discrimination embedded into these technologies.

Read Thematic Report

What’s Next

This project builds on earlier FSC-funded research from the Dais on AI’s impact on jobs and skills demand in Canada’s workforce. It forms part of a broader series of sector-specific deep dives into the implications of artificial intelligence for three sectors of interest: the public sector, the arts and culture sector, and the financial services sector.
Together, these projects will help policymakers, employers, and training providers anticipate the skills and workforce transitions AI will demand, and identify targeted strategies to ensure adoption strengthens both economic performance and social outcomes.