Six governments and ten AI firms sign Paris Call on frontier AI safety commitments

The declaration was published on November 9 after a video meeting hosted by French President Emmanuel Macron with about 50 participants. Signing governments are Bahrain, Canada, Finland, France, Norway and the Netherlands. Company signatories include Anthropic, OpenAI, Microsoft, Mistral AI, Cohere, Hugging Face, Black Forest Labs, Sakana AI, Domyn and Synthesia. The text commits signatories to disclose model capabilities, limitations and known risks, support evaluation by qualified independent evaluators, and share early warnings about serious incidents.

The Paris Call builds on the Seoul frontier safety commitments and on the Norway and Finland initiative launched in September. Critics, including Yoshua Bengio, say it lacks firm deadlines and mandatory independent audits before deployment. The document remains open to new signatories.

For companies building on frontier models, voluntary commitments like these tend to become standard questions in vendor reviews and enterprise procurement. Expect incident reporting and shutdown capability to appear in customer requirements, even though the pact has no enforcement mechanism.