Six governments and AI labs including Anthropic, OpenAI and Microsoft sign Paris Call for AI safety and security

Six governments and ten AI companies signed the Paris Call for AI safety and security on Monday, a declaration published by the French presidency after a call chaired by President Emmanuel Macron. The French government was joined by Canada, the Netherlands, Norway, Finland and Bahrain, while the company signatories include Anthropic, OpenAI, Microsoft, Mistral AI, Cohere, Hugging Face, Black Forest Labs, Sakana AI, Synthesia and Domyn.

Developers commit to disclose model capabilities, limitations and known risks, to support evaluation by qualified independent evaluators where proportionate, and to build fail-safe mechanisms that can shut down model operations within their control, including during training. They also pledge internal governance with real-time monitoring of emerging capabilities, red-teaming and reporting lines that protect internal researchers.

Signatories agree to publish non-sensitive evaluation findings in a standard format and to build a shared public repository that could hold red-teaming protocols, benchmarks and stress-test results. Governments commit to secure, confidential channels for whistleblowers and to work toward common incident classification and faster sharing of information on serious cyber and AI control incidents.

The text is a voluntary declaration rather than binding rules, and it acknowledges the risk of safeguards being removed from models while backing open-weight development. Civil society signatories include Yoshua Bengio's LawZero and SaferAI. The UK Prime Minister joined the call and said AI will top the agenda of the UK's 2027 G20 presidency, according to Downing Street, though the UK is not on the signatory list. The Élysée says the statement remains open to new endorsements.