An Inside Look into the Paris AI Summit

Last week’s Artificial Intelligence (AI) Action Summit in Paris brought together people from around the world to create a global AI agenda. From presidents and prime ministers to artists and members of civil society, attendees of this watershed event represented various points of view and motivations. 

France’s President Emmanuel Macron and his AI envoy Anne Bouverot kicked off the summit by emphasizing the need to prioritize inclusivity and consensus building in AI development. The two-day event culminated in 60 countries signing a declaration to ensure “AI is open, inclusive, transparent, ethical, safe, secure and trustworthy.” 

Inclusivity and consensus building, however, were accompanied by another dominating ethos: tech protectionism. The United States and the United Kingdom, for example, did not sign the declaration. Before leaving the summit, U.S. Vice President J. D. Vance criticized the European Union’s approach to regulating AI and emphasized that the United States would go its own way to achieve AI dominance.

To get an inside look into the discussions that took place during the event, Planetary Politics asked several members from civil society who attended or deeply analyzed the summit to share their insights. We spoke with three experts:

• Nicolás Grossman | Project Director, Global Index on Responsible AI, Global Center on AI Governance
• Laura Kupe | Co-founder and Principal, Congolese Diaspora Impact Strategies; Visiting Scholar, Planetary Politics, New America
• Faisal Lalani | Research Fellow, Global Center on AI Governance; Non-Resident Scholar, Planetary Politics, New America

The following takeaways have been edited for brevity and clarity.

Following the summit, some stakeholders have warned about the deprioritization of the ethical and human rights aspects of trust and safety on the AI Action Summit agenda. While this impression might be tempered by several important side events focused on these topics, this emerging trend can be linked to a range of factors, many of which became evident during the event. In an era of profound technological and societal transformation, some stakeholders see this moment as a unique window of opportunity to harness AI developments to reduce existing inequalities and better promote and protect human rights; for others, it represents a critical chance to prioritize the pursuit of new markets or geopolitical influence. These options are not mutually exclusive, but they effectively capture the range of positions expressed over the past few weeks and at the summit, especially considering recent political shifts among key actors.

Innovation and regulation should not be viewed as opposites, especially when it comes to impactful technologies like AI.

Just as societal development cannot be measured solely by GDP growth, AI innovation must be understood within its broader context and societal impacts. Global South actors face particular challenges: They need to leverage AI to tackle urgent developmental, economic, and social issues, while navigating the competitive landscape dominated by AI superpowers and big tech—all while mitigating internal inequalities and external power imbalances. This challenge cannot be addressed without international cooperation. Overall, my main takeaway from the AI Action Summit is that we missed the opportunity to address the deep global inequalities that AI is contributing to—inequalities that require collaborative international action.

I was part of the team which hosted “Trustworthy Innovation for a Connected World,” an official sideline event of the French AI Action Summit in Paris. CAS Strategies and Compiler organized an impactful gathering, highlighting how Global Majority countries are pioneering responsible and safe AI-enabled futures.

The event brought together government officials, technical experts, industry innovators, and civil society leaders to explore practical solutions and innovative approaches for building trustworthy, self-sustaining AI ecosystems that effectively address both global and local challenges.

A special highlight was the keynote address by the Minister of Technology of Barbados Marsha Caddle, whose insights underscored the transformative potential of AI when developed with inclusivity at the forefront.

The inclusion of “Trustworthy Innovation for a Connected World” as an official sideline event mirrored efforts at the highest levels of the summit, which focused on incorporating Global Majority perspectives in the development of a global AI agenda. The summit’s final declaration on “inclusive and sustainable” AI was endorsed by 60 countries, including several Global Majority countries like South Africa, India, Brazil, and Nigeria. I hope such collaborative approaches continue, positioning these nations as key architects in the evolving AI landscape. The world will be better for it.

AI has become a proxy for geopolitical power.

The summit certainly made sure to include the Majority World at the table. But in order to calcify these partnerships and make it sustainable, future declarations should offer more granularity around specific Majority World problems, present more commitments initiated by Majority World countries, and, in the long term, rebalance global governance structures of power to give Majority World countries proportional representation. 

Regulation is still being defined with Western metrics and Western ideals in mind; the Majority World is now taking steps toward crafting their own contextual paths to AI sovereignty. Perhaps it is time for the West to learn from the countries they’ve always seen on the periphery of technological prowess and realize they can balance both innovation and regulation.

The United States refused to sign the global commitment for safe and inclusive AI. The United Kingdom, likely fearful of losing U.S. engagement with their AI institutes, followed suit. India, in a continuation of its existing doctrine of digital control, cited its own digital public infrastructure as a model for responsible AI deployment while indirectly positioning themselves as representative of the Global South.

Perhaps most (pleasantly) surprising is that more countries did not refrain from signing the global commitment for safe and inclusive AI. It is emblematic of how aware most of the world is to the dangers of AI for existing public institutions and human rights. There is a real push here—a real opportunity to build cohesion where there has been fragmentation, persistence where overwhelming inevitability seems to loom, and inclusion where exclusion is seen as the only viable path to global leadership.