Middle-Power Nations Push Open-Source AI as Sovereignty Strategy

A coalition of middle-power nations is emerging around a shared strategy: building AI infrastructure on open-source foundations rather than relying on proprietary systems controlled by a handful of corporations.

Canada and the European Union have both released policy frameworks that place open-source AI at the center of their technological sovereignty plans. Canadian Prime Minister Mark Carney outlined the approach at this year's World Economic Forum in Davos, framing it as a way for nations to build technology that reflects values like human rights, sustainable development, and territorial integrity. Canada's national AI strategy explicitly prioritizes open-source development.

The EU followed with commitments to support open-source alternatives across the AI technology stack, fund startups working in the space, and revise government procurement guidelines to favor open-source innovation. Germany, Japan, and the United Kingdom have announced similar initiatives, with the UK launching an Open Source Builder's Fund aimed at attracting global talent.

Why it matters

The shift reflects growing unease about concentration of AI power and the geopolitical risks of dependence on foreign technology providers. When the U.S. government recently suspended access to a technology called Mythos, governments and companies worldwide confronted the reality that a single nation could unilaterally cut off access to critical infrastructure. Open-source models offer an alternative: technology that can be inspected, modified, and deployed without permission from a corporate or national gatekeeper.

The economic and technical case

The momentum isn't purely political. Research cited by Mozilla indicates that open-source technology has generated $8.8 trillion in demand-side value, with firms spending roughly one-third of what they would need to spend on software if open-source alternatives didn't exist.

Developer adoption is accelerating. A study from a16z and OpenRouter found that open-source models grew from 1–2% of token volume in late 2024 to nearly 30% by mid-2025. Entrepreneurs and researchers are building tools, models, and datasets tailored to local needs rather than defaulting to systems designed in Silicon Valley or Chinese AI labs.

Transparency as a design principle

Unlike proprietary AI systems, open-source models allow anyone to examine how they work. Mozilla, which updated its foundational Manifesto in December 2022 with a Pledge for a Healthy Internet, argues this transparency is essential for safety and accountability. The organization's four commitments center on inclusion, civil discourse, critical thinking, and collaboration across communities.

Mark Surman, writing for Mozilla, frames the question as one of values: whether AI will be owned and directed by a small number of actors or shaped more broadly in the public interest. The middle-power coalition represents a bet that open-source infrastructure can deliver both sovereignty and accountability.

These details were first reported by Fortune, based on commentary from Mozilla.