For the first time, the leaders of several artificial intelligence companies have signed the same text on AI safety. Announced by the Élysée, this signature concerns the “need for an international, multi-stakeholder mobilization to address the security risks raised by artificial intelligence”. Among the signatories are giants OpenAI, Anthropic and the French company Mistral.
The French presidency announced the signature on Monday, November 9, 2026. In its press release, it highlights three priorities: the need “to establish independent evaluators”, the need “to align on standards for evaluating and reporting major incidents”, and the need “to clarify the accountability of stakeholders”.
A coalition yet to be built
Speaking to franceinfo, Henry Papadatos, executive director of the French association SaferAI, welcomed an “important moment for AI safety and for French leadership”. He spoke alongside researcher Yoshua Bengio, whose expertise is recognized in the field of artificial intelligence.
For Henry Papadatos, the timeline is tight. He stressed that “AI is already accelerating its own development”, which in his view necessitates a rapid response. He calls for turning the three priorities into “concrete measures that a coalition of countries can take within a few months”.
Measures yet to be defined
The text signed by the companies sets out general principles, but it does not detail, in the elements reported by the source, the implementation modalities. The three priorities mentioned by the Élysée, particularly the creation of independent evaluators and the definition of a common framework for reporting major incidents, therefore remain to be concretized, according to statements gathered by franceinfo.
This initiative comes as the issue of artificial intelligence safety occupies an increasingly prominent place in public debate, without the source specifying the next steps in the process.
