Artificial intelligence (AI) is profoundly transforming our society, opening up unprecedented opportunities in many fields. For this technological revolution to benefit everyone, it is crucial to ensure its responsible and ethical development, with trust at its heart. AI raises major concerns regarding safety and security, as clearly demonstrated by the summits in Bletchley Park (United Kingdom - November 2023) and Seoul (South Korea - May 2024). Whether in anticipating extreme risks or addressing those already visible, a resolutely ambitious approach to building trust in AI is essential on an international scale.
Technological advances in AI also offer exceptional possibilities in the field of security. With this in mind, the AI Action Summit is committed to promoting safe and secure AI, particularly by providing the necessary tools to mitigate these risks.
For AI to fulfill its promises, a collaborative approach is essential. The AI Action Summit calls on public, private, and academic stakeholders to work together to build a trusted AI ecosystem.
This global approach is based on three pillars: science, solutions, and standards. With a robust international scientific consensus on AI, the time has come to develop technical solutions that are open and accessible to all, while creating common international standards recognized by the entire ecosystem. This will help prevent fragmentation and encourage convergence at all levels.
As part of the AI Action Summit, the Ministry of Europe and Foreign Affairs and the General Secretariat for Defence and National Security have supported the action of the thematic envoy, Guillaume Poupard, to federate an ecosystem of stakeholders, both nationally and internationally, mobilised to strengthen the safety and security of AI.