20 countries alongside the European Union call for increased international cooperation to ensure AI remains under human direction, supervision, and control.
The declaration proposes studying the establishment of an international organization with the power to set standards, support verification, and convene countries when AI exceeds certain capability thresholds.
The parties want to build common standards and share reports on serious AI safety incidents, while avoiding widening the technology access gap between countries.
Germany, South Africa, Canada, Australia, UAE, and Singapore are among the supporters, along with leaders from Norway, Kenya, Kazakhstan, Turkey, and the EU.
The US and China, the two leading AI superpowers, have not joined the declaration, while the text remains open for other countries to sign in support.
India, South Korea, Japan, the UK, and France are also not on the list of signatories mentioned in the article.
The move comes amid debates over AI safety after Anthropic CEO Dario Amodei called for slowing down development, receiving support from Sam Altman and Elon Musk.
OpenAI simultaneously proposed that the US lead in establishing international technical standards for advanced AI, amid growing concerns over AI-related cybersecurity incidents.
📌 Conclusion: AI governance is shifting from voluntary principles to discussions on an international mechanism capable of setting standards and supporting verification. The proposal by 20 countries and the EU emphasizes human control, incident reporting, and coordination when AI exceeds capability thresholds. However, the absence of the US, China, and several other technology superpowers shows that building a truly global AI governance framework still faces a major gap in interests and approaches.
