AI sovereignty emerged as an essential theme of the Paris AI Action Summit. Shortly before the summit started, President Emmanuel Macron stated that one of France’s strategic goals is to achieve technological sovereignty in AI.

Some of this isn’t new.

In 2021, France spearheaded efforts to establish “cloud sovereignty,” which eventually mandated stringent security and ownership restrictions on specific data centers owned by foreign companies in France. Since then – but especially since the public launch of Open Ai ChatGPT – France and the EU have attempted to enhance European AI capabilities through model development and compute infrastructure. Most recently, the EU announced its support for the OpenEuroLLM project, which aims to develop and deploy European open-source AI models trained primarily on European-owned supercomputers. 

However, two elements of Macron’s advocacy for European AI sovereignty look new and noteworthy.

First, Macron’s language is sharper, bringing a new tone and emphasis that suggests a clear and fast move toward his vision of AI sovereignty. Macron has stated that the future of AI is a political issue centered on sovereignty and strategic autonomy. He also underscored that Europe will compete globally and that its main competitors are the United States and China, though leaving the door open to closer collaboration with the US than with China.

Second, Macron’s idea of AI sovereignty isn’t one of France or Europe acting alone but of a strategic alliance among European and non-European countries – with India co-hosting the Paris AI Action Summit as a primary member of that alliance. Macron calls this a “third way” in AI development and deployment, a third pole of AI power to compete with and counterbalance the US and China.

This AI third way is not AI sovereignty in a traditional sense, which at a high level is a nation’s policy of placing the development, deployment, and control of AI models, infrastructure, and data in the hands of domestic actors. Instead, Macron’s third way is to establish an international network of nations that builds interdependence and mutual support among its participants, as well as independence and autonomy from the US and China.

A possible example of this AI third way is France’s announced answer to Stargate: a €109 billion grouping of French AI compute projects funded and supported by the United Arab Emirates, Canadian investors, and domestic companies like Orange and Thales (as well as US investors, suggesting that the third AI pole will be closer to the US than to China). 

Macron has also proposed joint efforts with India to pursue this third way. How strongly Prime Minister Narendra Modi will embrace the AI third way is an open question, and its strength will likely be heavily influenced by Modi’s meeting with President Donald Trump.

The French president’s laser focus on AI sovereignty reads as a response to recent US government actions on AI. Both the Joe Biden administration and the new Trump administration have promoted sovereign-like policies that encourage the development of AI in the US and reduce access to AI technology outside of the US. 

For example, the Biden administration’s AI diffusion rule, which the Trump administration has not revoked, would limit exports of advanced AI chips to most foreign countries. And while the Trump administration is still developing its AI action plan under the president’s January executive order on AI, Trump has already signaled strong support for domestic AI development.

The release of DeepSeek R1 also catalyzed France’s renewed effort to compete in AI via its national champion, Mistral, and by rallying its private sector to compete in AI. The Chinese model’s release has presented both an opportunity (showing that competition may come from less-resourced labs) and a threat (suggesting that models from China could further damage Europe’s AI competitiveness). 

France’s renewed call for AI sovereignty should motivate the US government – including the senior US government delegation in Europe this week and next –  to promote US leadership in AI, which it can most effectively (and maybe only) maintain and enhance through international AI collaboration.

American AI leadership and an embrace of collaboration are not mutually exclusive. The US should build on Vice President JD Vance’s speech at the Paris AI Action Summit to propose an alternative to the AI third way – one that includes the US as a principal partner, fosters collaborative AI investment and development projects, enables AI technology transfers in a manner that respects each partner nation’s self-determination, and addresses collective security risks (such as those potentially posed by DeepSeek R1) and other shared concerns. 

To do otherwise could turn America First into America Alone, ultimately damaging the AI leadership that the Trump administration seeks.

Pablo Chavez is an Adjunct Senior Fellow with the Center for a New American Security’s Technology and National Security Program and a technology policy expert. He has held public policy leadership positions at Google, LinkedIn, and Microsoft and has served as a senior staffer in the US Senate.

Bandwidth is CEPA’s online journal dedicated to advancing transatlantic cooperation on tech policy. All opinions expressed on Bandwidth are those of the author alone and may not represent those of the institutions they represent or the Center for European Policy Analysis. CEPA maintains a strict intellectual independence policy across all its projects and publications.