Middle Powers Forge New AI Alliances to Challenge US-China Dominance

As the United States and China continue their fierce competition for AI supremacy, a coalition of middle powers—including Canada, Germany, France, Japan, South Korea, Israel, and the UAE—is emerging with a bold strategy: developing sovereign AI systems that prioritize local innovation, infrastructure, and data control. These efforts aim to provide viable alternatives to the dominant US and Chinese AI technologies.

This new geopolitical landscape is exemplified by a recent transatlantic partnership between Canadian AI firm Cohere and German startup Aleph Alpha, backed by a $600 million investment from Schwarz Group. Their collaboration highlights the urgency and ambition of these nations to carve out a meaningful role in the global AI ecosystem, ensuring they are not sidelined in the race for AI-driven economic and strategic power.

Transatlantic Alliance: Cohere and Aleph Alpha Join Forces

Canadian AI company Cohere, founded by researchers with roots at Google Brain, has struggled to keep pace with US giants like OpenAI and Google. To strengthen their position, Cohere has partnered with Germany’s Aleph Alpha, forming a strategic alliance supported by a $600 million commitment from Schwarz Group. This partnership aims to pool resources and expertise to compete more effectively on the global stage.

Europe’s Push for AI Sovereignty: The Mistral AI Example

Paris-based Mistral AI is positioning itself as a distinctly European alternative to American and Chinese AI providers. Emphasizing control over AI infrastructure and data, Mistral offers 'open-weight' models that allow users to customize and deploy AI systems independently, even offline. This approach appeals to governments and enterprises wary of opaque, centralized AI systems.

"AI should be a tool for empowerment, not dominance."—Arthur Mensch, CEO of Mistral AI

Mistral’s strategy has attracted significant investment, including a $2 billion funding round led by ASML, Europe’s most valuable tech company, which also plans to integrate Mistral’s AI into its products and research.

Expanding the Middle-Power AI Ecosystem

Beyond Canada and Europe, other middle powers are ramping up AI initiatives. Japan’s Sakana AI is building domestic AI infrastructure, South Korea is launching a large state-led AI program, Israel’s AI21 Labs is exploring strategic growth options, and the UAE’s G42 is heavily investing in AI compute capabilities. Additional European players like Germany’s DeepL and the Netherlands’ Axelera AI further diversify this growing ecosystem.

Challenges and the Road Ahead for AI Middle Powers

According to Anton Leicht of the Carnegie Endowment for International Peace, middle powers face a complex dilemma: they must find niches to participate in AI progress, secure access to compute resources largely controlled by the US and China, and protect themselves from AI-driven threats by leveraging foreign defensive technologies.

By joining forces, companies like Cohere and Aleph Alpha are attempting to overcome these hurdles. While their success is uncertain, their efforts underscore a critical goal for middle powers—not to win the AI race outright, but to ensure they remain active players in shaping the future of AI.