Research Highlight
Artificial Intelligence
Rural Governance
Policy
📅 April 2026
📚 Brandon University & Scotland's Rural College
🌎 Canada & Scotland
A new evidence brief from researchers in Canada and Scotland asks whether artificial intelligence can — and should — play a role in how rural communities are governed. The answer, it turns out, is complicated.
Governments on both sides of the Atlantic are facing two challenges at once: crafting rural policies that keep pace with changing realities on the ground, and responding to the rapid rise of AI in public decision-making. A collaboration between Brandon University in Canada and Scotland's Rural College set out to examine both — and to ask whether rural places are even part of the conversation.
Reviewing over 220 pieces of academic and grey literature from the UK, Scotland, and Canada — including policy documents, government reports, blog posts, and independently commissioned research — the team found significant gaps and emerging risks that policymakers would do well to heed.
What the research found
|
No shared definition
There is no agreed understanding of what "AI" actually means across different government contexts — political, economic, or social.
|
Promising use cases
Machine learning and predictive analytics show real promise in health, environmental management, and agriculture.
|
Rural voices missing
Rural communities are largely absent from AI strategy discussions in both Canada and Scotland.
|
Fragmented governance
AI governance in Canada remains patchy. Scotland has a national AI strategy; the UK's appears shaped largely by a single 2025 report.
|
The challenges facing rural areas
The research identifies a cluster of issues specific to rural and remote communities that current AI governance strategies largely ignore:
► Authority and accountability — Private sector actors, particularly technology companies, appear to be exerting outsized influence over how AI governance frameworks are shaped, raising questions about who is responsible when things go wrong.
► Biased data, harmful outcomes — AI systems built on incomplete or skewed data risk amplifying existing inequalities. Marginalised, rural, remote, northern, and Indigenous populations are especially vulnerable.
► The rural digital gap — Many rural areas face data poverty, shortages of digital skills and expertise, tight budgets, and outdated infrastructure — all significant barriers to meaningful AI adoption.
► No opt-out — There are very few mechanisms for communities that do not wish to be subject to AI-driven decision-making.
► Short-termism — Current approaches prioritise cost savings and visible results over long-term public benefit, ethics, or structural change.
"AI systems mirror societal biases and may worsen inequalities for marginalised, rural, remote, northern and/or Indigenous populations."
Kelly et al., SSHRC Evidence Brief, 2026
What needs to change
The researchers are clear that more evidence is urgently needed — not just about the promises of AI, but about how it actually performs in real policy settings, particularly rural ones. Key recommendations include:
► Developing clear, place-based AI governance frameworks that account for the specific capacity, infrastructure, and needs of rural communities.
► Establishing transparent accountability mechanisms, including clear lines of responsibility and options for redress.
► Ensuring that public institutions are not captured by private technology interests, and that AI adoption is guided by principled, evidence-informed oversight.
► Treating rural communities not as an afterthought, but as active participants in shaping how AI is used in governance.
📄 About this research
This post is based on an evidence brief by Wayne Kelly, Shirlyn Kunaratnam, S. Ashleigh Weeden, Nicole Breedon, Ian Merrell, and Joel Templeman. Published by Canada's Social Sciences and Humanities Research Council (SSHRC) in April 2026. The full report is forthcoming.
→ Read the original evidence brief on the SSHRC website
