Evidence for Democracy Examines Growing Role of Artificial Intelligence in Policymaking
OTTAWA — A new report from Evidence for Democracy is raising questions about transparency and accountability as artificial intelligence tools become increasingly embedded in Canada’s federal public service.
Titled “AI and Democracy: Navigating Trust, Truth, and Technology in Policymaking”, the report provides what the organization describes as the first comprehensive overview of how AI systems are being deployed across federal departments and agencies, and the potential implications for democratic governance.
The findings suggest that while AI adoption is accelerating within government operations, oversight and public engagement have not kept pace—prompting calls for a broader national conversation about the role of emerging technologies in decision-making.
Growing Use of AI Across Key Government Functions
According to the report, AI tools are now widely used across multiple sectors of the federal government, with governance, security, and economic functions representing the most prominent areas of deployment.
The Canada School of Public Service leads all federal entities in AI adoption, with 32 systems currently in use. It is followed by Innovation, Science and Economic Development Canada with 27 systems and Statistics Canada with 26.
The report also highlights the extent to which external vendors are involved in building these systems. Two out of five AI tools currently in use were developed by third parties, with a significant majority of those vendors based in the United States. Nearly one-third of externally developed systems are linked to three major technology companies: Microsoft, Google, and Amazon.
This reliance on foreign-owned technology providers raises additional questions about data sovereignty, procurement practices, and long-term strategic control over critical digital infrastructure.
AI-Generated Content Playing Expanding Role
One of the report’s most notable findings is the increasing use of AI systems to generate content that directly influences federal decision-making processes.
Researchers found that approximately one in five AI tools is being used for this purpose, including the creation of AI-generated videos and voiceovers, large language model (LLM)-drafted police reports, and AI-assisted policy briefs.
Such applications, the report suggests, represent a shift from administrative or analytical uses of AI toward more substantive roles in shaping policy outcomes.
“When AI systems are generating content that shapes federal decision-making without adequate transparency or public consultation, we’re fundamentally altering the relationship between government and citizens, and we need to have this discussion now,” says Félix Proulx-Giraldeau, Interim Executive Director at Evidence for Democracy.
Trust, Accountability and Democratic Risks
The report situates its findings within a broader context of declining public trust in democratic institutions and the growing influence of digital technologies on information ecosystems.
It argues that the integration of AI into policymaking processes—particularly without clear disclosure or accountability mechanisms—could further erode public confidence if left unaddressed.
“At a time when Canada’s democracy is facing historic lows in public trust, coordinated misinformation campaigns, and major disruptions posed by new AI systems and technologies, we need to take action to safeguard our democracy and ensure that evidence-informed decision-making continues to build a better future for all Canadians”, says Trevor Potts, Director of Research and Policy at Evidence for Democracy.
Call for Policy and Public Engagement
The report calls for stronger governance frameworks, increased transparency around AI use, and more robust public consultation as federal institutions continue to adopt these technologies.
For policymakers, the findings underscore the need to balance innovation with safeguards that protect democratic norms. For industry stakeholders, the report signals growing scrutiny of how AI systems are developed, procured, and deployed within government contexts.
As Canada continues to position itself as a leader in artificial intelligence research and development, the report suggests that ensuring public trust will be critical to sustaining that leadership.
The release of “AI and Democracy: Navigating Trust, Truth, and Technology in Policymaking” adds to a growing body of research examining the intersection of technology and governance, and is likely to inform ongoing discussions about regulation, ethics, and the future of AI in the public sector.