\n<\/p>\n
UNITED NATIONS — <\/span>The heads of major artificial intelligence firms pleaded with the United Nations on Wednesday to save the world or at least its people \u2014 by somehow regulating the fast-expanding technology<\/a> that they have been designing.<\/p>\n \u201cIf managed poorly, I even believe AI could be a risk to humanity as a whole,\u201d said Dario Amodei, chief executive officer of Anthropic. And from his competitor Sam Altman, CEO of OpenAI, came this assessment: \u201cWe could lose control of the future to AI.\u201d<\/p>\n Both said the countries of the world, through the U.N. Security Council, where they testified Wednesday, have to set safeguards of some kind to prevent the technology from getting too powerful to rein in.<\/a><\/p>\n They also called for the world to make sure that the power of AI isn\u2019t concentrated in one company or country. They also touted the benefits of the technology for bettering humanity and what Altman called \u201cgiving people more power over their own lives.\u201d<\/p>\n Those tech leaders \u201care telling us in stark terms that it cannot be left to them to prevent disaster from befalling the world,\u201d British Foreign Secretary Ed Miliband said at the Security Council. \u201cWe cannot ignore them because they are right.\u2019\u2019<\/p>\n Miliband said when the U.K. leads the G20 next year, it will put controlling AI at the heart of discussions, looking for a single set of global standards \u201cto insure that AI development is safe.\u201d<\/p>\n But, the United States said: Don\u2019t do it.<\/p>\n Worries about control of AI are \u201cnot a reason to pause its further development or to constrain it with new global governance structure,\u201d said White House Science Adviser Michael Kratsios. \u201cWe want you, our allies and partners, to share in the benefits of this technology.\u201d<\/p>\n The U.S. and China are jostling for AI dominance \u2014 and Chinese U.N. Ambassador Fu Cong emphasized that all countries should be able to choose AI technology as they see fit.<\/p>\n He warned against \u201cbanding together into a petty us-vs.-them clique in the tech sector and forcing other countries to take sides\u201d and enabling unnamed major powers that he described as seeking \u201cdigital hegemony.\u201d He called on countries instead to promote open-source models — China\u2019s approach<\/a> to the technology.<\/p>\n For most of the world leaders, one key question about AI is this: Who should control life-and-death decisions \u2014 human or machine?<\/p>\n Amodei, Altman and Hugging Face CEO Cl\u00e9ment Delangue, who spoke via a glitchy remote connection, spoke with the U.N.\u2019s most powerful council because they and world leaders are trying to figure out not just whether to control AI, but by whom, how it could possibly be done and when it might be just too late. <\/p>\n The day before, U.N. Secretary-General Antonio Guterres warned of \u201ckiller robots.\u201d Experts and European leaders said that\u2019s not hyperbole but something that\u2019s either already a grim reality or perilously close.<\/p>\n Amodei called on world leaders to agree on not letting AI build biological weapons, figure out how to verify what models have the capability to do and \u201cestablish common standards\u201d for testing and for loss of control and misuse.<\/p>\n French Foreign Minister Jean-No\u00ebl Barrot, whose country convened Wednesday\u2019s meeting, compared the current state of AI to both the runaway computer HAL9000 in the movie \u201c2001: A Space Odyssey\u201d and the dawn of the atomic age.<\/p>\n \u201cJust as with the atom back then, the international community must now regulate AI in order to make the most of its potential and avoid the worst,\u201d Barrot said.<\/p>\n His wasn’t the only concern expressed during this high-profile U.N. meeting week. <\/p>\n \u201cWe must not allow the cost of today\u2019s hesitation to become tomorrow\u2019s catastrophe,\u201d said Liberian UN Ambassador Lewis G. Brown II.<\/p>\n Ukrainian President President Volodymyr Zelenskyy, speaking at the General Assembly, said some kind of human control over machines must achieved soon. Otherwise, he said, for his country \u2014 at war for 4\u00bd years \u2014 it may be just too late.<\/p>\n \u201cThere is already a real possibility that AI, not only people, will begin to decide what happens on the battlefield. And we need peace before we reach that point,\u201d Zelenskyy said.<\/p>\n Last month, a Russian drone seemingly using AI \u201cautonomously made a targeting decision that killed three Ukrainians. Killer robots are technically easy to build, and they are real. We should not be under any illusions otherwise,\u201d Kentaro Toyama, an information technology professor at the University of Michigan, said in an email.<\/p>\n Data scientist Rumman Chowdhury, co-founder of the accountability nonprofit Humane Intelligence, said the idea of killer robots \u201cevokes a science-fiction villain, when the question I want the public to ask is much more concrete: Who authorized a machine to select and attack a target, under what constraints, and who answers for the result?\u201d<\/p>\n The U.N. meeting unfolded as Chinese President Xi Jinping<\/a> headed to Washington for a much-anticipated state visit. AI interests are looking to<\/a> what comes out of Xi\u2019s discussions with U.S. President Donald Trump. Earlier, U.S. Treasury Secretary Scott Bessent said the two countries had discussed a potential new \u201cnotification mechanism\u201d<\/a> for AI incidents that could impact national security.<\/p>\n But on Tuesday, Trump, lauding AI as \u201csuperintelligence,\u201d said the United States would reject efforts to control the technology<\/a> as it tries to win the development race with China. Kratsios was the one of the few who used that term in the Security Council.<\/p>\n Meanwhile, regional groups in Europe, Asia and elsewhere also are trying to coordinate AI policies, and various nations \u2014 and even subcomponents, such as U.S. states \u2014 are crafting their own. <\/p>\n ___<\/p>\n Farnoush Amiri, Edith M. Lederer and Jennifer McDermott contributed to this report from the United Nations.<\/p>\n ___<\/p>\n The Associated Press\u2019 climate and environmental coverage receives financial support from multiple private foundations. AP is solely responsible for all content. Find AP\u2019s standards<\/a> for working with philanthropies, a list of supporters and funded coverage areas at AP.org<\/a>.<\/p>\n<\/div>\n