As the extensive harms and risks of artificial intelligence enter mainstream public debate, leaders are being forced to acknowledge how quickly the technology has advanced relative to their capacity to respond. For years, a handful of increasingly powerful men, concentrated in two countries, have been allowed to set the terms for artificial intelligence. 

Some of these men publicly acknowledge some of the risks of allowing the unfettered development of this powerful technology. Others do not.  

Regardless, it must not be left to them alone to decide how AI is governed. As emphasised by The Elders’ in their new AI Principles for Leaders, it is a government’s primary responsibility to protect its people from harm and potential harm. And no government can do so by acting alone, given the global reach of AI, and the significant gaps in technological and institutional capacity between countries.  

Equally, governments cannot govern AI if they do not reassert their independence from AI companies, and enforce transparency and accountability. Governments are responsible for subjecting all technologies to the law – AI included.  

Crucially, leaders need to regain the trust of their people on this critical issue. This requires telling the truth about AI’s harms and risks; prioritising protection of the environment, livelihoods, and human rights over profit; and providing meaningful opportunities for people to steer AI in a direction that benefits all, not a few wealthy individuals. 

Nobody should trust a government willing to use weapons that select their own targets, especially when those decisions are informed by a chatbot. Neither should they trust a government that abdicates its responsibility to manage border enforcement, policing, or social security to AI systems.  

Yet this is the reality we are now faced with.   

Leaders must adopt a decisive and principled approach if they are to change the trajectory of AI so that it becomes a force for good, for all countries and communities. We are heartened by the Call for Control of Frontier AI Models that 30 countries and counting have signed onto. But this is only a first step.  

We, as a group of AI researchers, academics, and policy-makers, welcome The Elders’ AI Principles for Leaders, which ask leaders to prioritise humanity, safety, cooperation, and long-term thinking, among others. We encourage all who lead – in particular those with the greatest influence – to urgently act by them.  

Stuart Russell, OBE, FRS, Distinguished Professor of Computer Science, University of California, Berkeley, and President, International Association for Safe and Ethical Artificial Intelligence 

Marietje Schaake, Non-resident fellow at Stanford, Author of The Tech Coup 

Cynthia Dwork, Gordon McKay Professor of Computer Science at Harvard University, US National Medal of Science Laureate, NAS, NAE 

Alondra Nelson, Harold F. Linder Professor, Institute for Advanced Study; former Acting Director, White House Office of Science and Technology Policy 

Kate Crawford, Professor at the University of Southern California, member of AI Council of Prime Minister Sánchez of Spain, author of Atlas of AI. 

Lucy Suchman, Professor Emerita, Lancaster University UK, Anthropologist of Science and Technology, Benjamin Franklin Medal in Computer and Cognitive Science 

Philip Thigo, Special Envoy on Technology, Republic of Kenya 

Dame Wendy Hall FRS, FREng, Regius Professor of Computer Science, University of Southampton 

Yi Zeng, Professor, Gaoling School of AI, Renmin University of China, and Founding Dean of Beijing Institute of AI Safety and Governance