New Delhi: Twenty countries and the European Union have called for international cooperation to keep Artificial Intelligence, or AI, under human control, including forming a global body to set and enforce uniform standards.
The declaration, titled “A Call for Control of Frontier AI Models,” was launched on 21 September 2026, on the sidelines of the United Nations General Assembly (UNGA) in New York. It was released by Finnish President Alexander Stubb and Norwegian Prime Minister Jonas Gahr Støre.
The joint statement shed light on AI’s potential to improve lives and advance science while warning that rapid advancements pose “serious” safety and security risks.
“Recently we have seen capable AI systems circumventing testing safeguards, exploiting vulnerabilities and gaining unauthorised access to real-world systems,” states the declaration. “Leading scientists and executives are warning that the pace of development could outpace our ability to manage emerging risks.”
A total of 22 political leaders from 20 countries and the European Union have signed the declaration. The countries are Norway, Finland, Australia, Bahrain, Canada, Denmark, Estonia, Germany, Iceland, Ireland, Kazakhstan, Kenya, Latvia, Moldova, the Netherlands, Singapore, Spain, South Africa, Turkey, the United Arab Emirates, and the European Commission.
The official document remains open for endorsement by additional countries.
“AI represents one of the most profound technological changes humanity has ever experienced. It will affect science, economics, politics, security and what it means to be human. Humans should be in the driver’s seat in this shift. We have choices in how we respond to AI: we can compete, cooperate, or step back. The right choice is to cooperate,” Stubb said.
What would the watchdog do?
According to the declaration, the proposed UN-backed global AI watchdog—modelled conceptually after the International Atomic Energy Agency (IAEA)—would have three primary functions:
1) It should keep a check on whether companies have developed transparent safety protocols. These protocols should include mandatory pre-deployment testing and independent evaluations, with qualified evaluators granted sufficient access to assess risks.
2) It should facilitate inspections and verifications whether ensuring that “governments and regional organisations further develop and coordinate common standards.” It must strengthen transparency—including the shared reporting of serious safety incidents—and ensure that countries across all regions have equitable access to scientific capacity, expertise, and trusted evaluation.
3) The body should be able to set standards, convene states and enable verification.
“The development of frontier AI models must be brought under better control. AI is an extremely powerful technology with vast potential, but it may also pose a serious risk to safety and security. Addressing these risks requires political action. We are working closely with other countries and will actively promote this issue during the UN High-level Week,” said Støre.
Also read: Why UK lawmakers want Buckingham Palace guards to ditch bearskin caps
Who have not signed as of now?
The declaration has not been signed by major powers, including the US, China, and the UK.
US President Donald Trump and his Chinese counterpart Xi Jinping are holding a high-stakes, three-day summit in Washington DC, marking Xi’s first visit to the US capital in over a decade. The duo is expected to discuss AI on Thursday. Top tech executives like Sam Altman (OpenAI), Jensen Huang (Nvidia), Elon Musk (Tesla), and Sundar Pichai (Alphabet) have been invited to the event.
Both Washington and Beijing think of AI as a critical driver of future economic and military growth. While international watchdogs and safety advocates warn of the catastrophic risks of unchecked AI, both nations are locked in a security dilemma. Neither superpower is willing to slow down its development out of fear that the other will surge ahead and gain an advantage.
Meanwhile, a few other world powers which are yet to support the declaration are India, South Korea, Japan, and France.
Also read: Krishi Bhawan solved India’s food crisis; Shastri Bhawan saw a young democracy grow
‘Catastrophic risks’
The declaration comes at a time when tech giants are demanding a pause on advanced development to prevent “catastrophic risks”.
In a rare unity, Elon Musk and Sam Altman backed Dario Amodei, asking to ‘slow the pace’ of AI development.
“I continue to believe that AI can enormously improve the quality of human life. But the benefits will only be achieved if we build the technology in the right way, and — so long as we use the time we gain well — it is worth taking unusually deliberate care to get it right. The measures I propose to advance the frontier at a safe pace will not be easy. But I believe we owe it to humanity to try,” Amodei had written in an essay.
Musk backed him and shared his post on X. “Dario is right,” he had posted.
He had been raising concerns about AI safety since 2017.
“If you’re not concerned about AI safety, you should be. Vastly more risk than North Korea,” Musk posted on X in 2017.
On Thursday, Musk is expected to bring these safety arguments to the White House, where he will join other tech elites at the formal state dinner for Xi.