The UK AI Security Institute currently tests major AI models on a voluntary basis, and AI developer Anthropic reportedly refused to submit its latest model for testing.

Overview Debates Related claims Topics Sources

A declaration proposed by Norway calling for AI to remain under human control was signed by more than 20 countries, but the United Kingdom, the United States, and China did not sign the declaration

UK Prime Minister Andy Burnham warned that artificial intelligence will act as a "multiplier of risk" and increase the threat of state-sponsored disinformation campaigns.

U.S. Treasury Secretary Scott Bessent and Chinese Vice Premier He Lifeng met in New York on September 20, 2026, to discuss trade, tariffs, and artificial intelligence safety.

British Prime Minister Andy Burnham expressed a desire to establish a single set of global principles and standards for artificial intelligence, which contrasts with U.S. President Donald Trump's rejection of globalist schemes to control the technology.

The U.S.-China AI dialogue proposed by Treasury Secretary Scott Bessent includes an emergency notification mechanism to communicate about AI incidents that rise to a national security level

Anthropic CEO Dario Amodei published an essay on September 12, calling for leading AI labs to voluntarily slow down the pace of capability development

The agenda for the Trump-Xi summit in Washington on September 23, 2026 includes artificial intelligence safety, the Pentagon's 1260H blacklist, trade tariffs, and geopolitical issues like the war in Iran

OpenAI CEO Sam Altman and SpaceXAI CEO Elon Musk publicly endorsed Anthropic CEO Dario Amodei's proposal for a coordinated slowdown in AI development, from his essay 'We Must Pace the Frontier.'

US Representative Ro Khanna stated on September 20, 2026, that Donald Trump and Xi Jinping must reach an agreement banning recursive self-improving AI and establishing safeguards against AI misuse for biological or nuclear weapons.

The artificial intelligence company Anthropic blocked accounts in Houthi-controlled northern Yemen that were using its Claude AI model to develop software for guiding and stabilizing rockets