Global cooperation is urgently needed to manage the growing risks of artificial intelligence, experts have warned, as major powers compete for dominance in the rapidly advancing technology.
The concerns have intensified amid the rivalry between the United States and China. US President Donald Trump has argued that America must win the global AI race, saying the country that leads the technology could gain a decisive strategic advantage.
Critics, however, warn that treating AI development purely as a geopolitical contest could create risks that no single country would be capable of containing. They argue that the consequences of advanced AI could extend far beyond national borders.
AI systems have largely operated under human supervision, but increasingly autonomous systems are raising questions about how long meaningful human control can be maintained. Experts have identified possible threats involving cybersecurity, biotechnology, disinformation and military escalation.
The development of autonomous weapons has also raised concerns over accountability and human decision-making in life-and-death situations. Experts say reducing the risk of AI-related extinction should be treated as a global priority alongside pandemics and nuclear war.
Advanced AI systems can reason, plan and perform increasingly complex tasks, but they lack human empathy. Without effective safeguards, experts warn they could contribute to cyber conflicts, facilitate access to dangerous technologies or enable manipulation on an unprecedented scale.
Several international initiatives have already called for stronger global oversight of AI. In May 2023, the Elders urged world leaders to establish international governance mechanisms that would maximise AI’s benefits while limiting its dangers.
The Bletchley Declaration, adopted at the AI Safety Summit in November 2023, similarly recognised that many AI risks cross national borders. It therefore called for international cooperation to address the challenges posed by advanced artificial intelligence.
The United Nations has also taken steps towards greater AI oversight by establishing an Independent Scientific Panel on AI and a Global Dialogue on AI Governance. The UN Security Council has separately examined AI’s implications for international peace and security.
Despite increasing strategic tensions between Washington and Beijing, experts say both countries have an interest in preventing the most dangerous consequences of advanced AI. The US and China are widely regarded as the world’s two leading AI powers.
Both countries are reportedly considering ways to contain extreme AI risks and prevent powerful systems from falling into malicious hands. Chinese President Xi Jinping has also called for measures to prevent and manage AI-related risks.
The Rome Declaration on disarming AI, issued by the Global Nobel Laureates Assembly on Artificial Intelligence and Nuclear War, has added to international calls for greater restraint. Supporters say such initiatives demonstrate the growing recognition that AI safety requires cooperation.
Experts point to previous global challenges as evidence that strategic rivals can cooperate when facing threats that affect humanity as a whole. Countries with competing interests have previously worked together on nuclear proliferation, climate change, infectious diseases and ozone-layer depletion.
More recently, nations have reached agreements on issues including cybercrime and ocean conservation. These examples suggest that shared threats can create opportunities for cooperation even when countries remain divided on wider political and strategic issues.
The Strategic Foresight Group’s report, The Essential Convergence, has proposed a Global Compact on Extreme AI Risk. Rather than creating one comprehensive global regulatory system, the proposal calls for a coordinated framework that countries can adopt while retaining flexibility in their domestic AI policies.
Under the proposed framework, countries could agree on common definitions of extreme AI risks, minimum safety standards and collective response mechanisms. It would also seek to establish systems capable of operating across national borders.
Such an arrangement could create a minimum global safety baseline while allowing countries to continue developing AI for economic, scientific and other beneficial purposes. The rapid advancement of the technology offers enormous opportunities, but its risks could equally transcend national boundaries.
Experts therefore argue that international cooperation must keep pace with AI development. With the US and China competing for technological supremacy, the immediate challenge is to identify areas where strategic rivals can agree on basic safeguards.
The central argument is that AI development does not necessarily have to stop. However, efforts to prevent its most catastrophic consequences cannot be postponed until the risks become irreversible, making international cooperation an urgent priority.
Comments 1