Speech by the President of the Government of Spain at the opening of the first meeting of the United Nations Panel of Experts on AI

First and foremost, I would like to thank the Speaker of Congress for hosting this important multilateral meeting of the Independent International Scientific Panel, which will help us all-all of society, and governments-to untangle, comprehend and navigate the challenge of the technological revolution. So many thanks, Madam Speaker, for hosting us in this room, the Constitutional Room, one of the most important spaces of Spain's Congress of Deputies.

I must say that I am very happy- very happy-and very grateful to you all for the effort you have made to come to Madrid, to hold your first in-person meeting in this city. Madrid is always a good place to visit. I am also very happy because, firstly, I think with this event we are strengthening the multilateral system, which is under attack, and we are doing so in an area-artificial intelligence-where a global multilateral response is clearly needed, in which all nations feel that they are on an equal footing; also, with an alliance of businesses, academia, the scientific community, legislatures and executives. That is why we in Madrid thought it would be very fitting to hold this meeting in the building at the heart of our country's national sovereignty.

The issues have been addressed by the experts in much more eloquent terms than my own, and also by the Speaker of Congress, but I would like to offer some thoughts and ideas nonetheless. It is a shame that this meeting will probably not be the top story on the news, or in the headlines of the papers either today or tomorrow, but it remains a landmark in terms of the commitment from Spain, and from the multilateral system, to addressing major challenges such as the development of artificial intelligence.

As I mentioned, it has already been said that artificial intelligence poses a lot of questions, because the technology can be used for a variety of different purposes. It can be used to save lives, as is already happening in hospitals, or to end lives, as is sadly already happening in theatres of war not too far from here. With so much at stake, it is only right that we ask ourselves who is setting the course for artificial intelligence, who is choosing the path ahead for the most transformative technology of our times. I think we can all agree that the answer is clear, but also disconcerting. Needless to say, the decisions are not being made by societies; or by countries' parliaments; or by democratically elected governments; not even by the market; instead, the decisions are being made by a handful of companies. Suffice to say, globally just five companies account for almost 60% of the generative artificial intelligence market-60% controlled by just five companies-and all five are from the same country, which is home to 74% of global compute capacity for artificial intelligence, while the European Union accounts for-as the representative of the University of Santiago de Compostela said-just 5% of that capacity.

In effect, that silent colonialism is happening around the world, including Europe. So we are not talking about a concentration of capacity, but rather a concentration of power, because technology is power. What we are witnessing is not the formation of a market, but the formation of an oligarchy that only serves its own interests. What is more, just to give you an idea of the scale, in 2025 that country's private sector invested 286 billion dollars in artificial intelligence-in 2025, a total of 286 billion dollars invested in artificial intelligence in the United States-equivalent to 17% of Spain's gross domestic product, 14 times more than Europe and 23 times more than China.

I would also like to raise another question: of all that investment, how much went into security to protect against artificial intelligence developments? Because I am sure that amount was minuscule. So I believe we have an enormous amount of work to do. The question we must ask ourselves is who is going to benefit from this technology and how. That is why I think it is very important for all nations to be on an equal footing, and for it to be the United Nations that is hosting, convening and coalescing this debate.

Many are convinced that technology on its own can bring about social progress. That is only partially true. It is people who truly build social progress. Therefore societies and the executives that legitimately and democratically represent societies-alongside legislatures-are the ones that ensure that technological progress does not cause regression in other areas. There are multiple examples of this throughout history, and the Speaker of Congress mentioned some earlier. The steam engine undeniably drove the industrial revolution, but it also gave rise to labour exploitation, inhumane working hours and child labour. Nuclear power launched a new era for science, indubitably, but also caused devastation in Hiroshima and Nagasaki. Social networks promised a more open, connected and democratic society, but as María rightly said earlier they are now a "failed State", with ever more resources and financial power to damage and undermine peaceful coexistence and democratic systems.

It was society as a whole-scientists, academics, trade unionists, businesses, the middle class and also the legislatures and executives-that turned the industrial revolution into improved well-being. It was governments and the scientific community that limited the use of nuclear power. Regrettably, that nuclear non-proliferation treaty has now expired, and Spain is advocating that the nuclear powers concerned sit down at the negotiating table to agree a new treaty on non-proliferation of nuclear weapons. Therefore, it must also be the public powers, alongside businesses, academia, the scientific community and representatives of society that turn the digital space into what we were promised.

Clearly no technology is neutral. It depends on who controls it, how it is regulated and what it is used for. I believe that with artificial intelligence we are reaching a historic turning point, a key juncture at which our actions will unquestionably be judged by the future generations mentioned earlier. This is no run-of-the-mill technological advancement. It is a technology with broad-reaching and unforeseeable impacts, with the potential to bring incredible benefits, but also to exacerbate risks and pose unprecedented threats to humankind.

It is true that we have advanced, advanced a great deal from a technological perspective; what I am less sure of is whether we have made equal progress on the more moral aspects of how to govern and how to channel this artificial intelligence. We have been hearing much talk of rearmament recently; I believe we must also rearm ourselves morally if we are to rise to the challenges posed by the technological advances brought by human expertise, innovation and science.

I believe that AI must, in all cases, be governed by common rules for everyone, based on shared principles, ethical and democratic accountability and effective multilateralism. That is, I believe, the vision of the European Union. It is this vision too, of course, that Spain is respectfully trying to offer. Under the Spanish Presidency of the Council of the European Union, Europe took the lead in establishing a common framework on artificial intelligence. The purpose is not to stand in the way of innovation-that would be impossible, there is no holding back the tide-but to at least channel it by applying parameters and a framework with a human-centred and moral approach to the challenges posed by artificial intelligence, which we must face together. To ensure that the most transformative technology of our times advances in the same direction as our societies and our democracies, with safeguarded rights, with legal certainty and with peace of mind.

Let us be clear-not even the creators of the most powerful models themselves believe that these models can be governed by market forces alone. We must therefore, all of us, do our part to ensure that future generations will see that we rose to the challenge this represents. We must do so with the courage of our convictions, decisively, without fear.

A hugely important step was taken in 2024, with the assessment of the situation from the Secretary-General of the United Nations, António Guterres, in the report on governance of artificial intelligence. That same year, as you so rightly mentioned earlier, 193 States, Spain among them, adopted the Global Digital Pact as a framework for action and the General Assembly created this Scientific Panel and Global Dialogue, the most ambitious such architecture created to date; the most ambitious, that is why I am so proud that Spain and this parliament are hosting the first step of your journey-this architecture, the most ambitious that exists today, to govern artificial intelligence, as I said before, in a multilateral, collective undertaking.

Spain, I want you to know, shares this vision, as do many other governments, many other nations, because it is a matter of common sense. That is why it is an honour for our country to host this first meeting of the Scientific Panel, especially in the Congress of Deputies. We believe in diligence, in the value of multidisciplinary experience. It is our position that the governance of artificial intelligence must be in the hands of the people, and not a tiny group of individuals. We-and by we, I mean societies, but also governments-do not want to be mere observers of this technology. We also want to be involved in designing its architecture.

This is why, together with the United Nations, we have launched the AI Governance for Humanity Lab, a forum for governments and companies to jointly define the standards for ethical and trustworthy AI. I am among those who believe-and I understand that many of you here today share this position-that, of course, national regulation is important, and it goes without saying that it is important at the European level. I hope, too, that we can achieve it at the multilateral level. But without the backing of businesses, without the participation of the scientific community, of academia and of society as a whole, it is going to be very difficult for regulation alone to enable us to channel artificial intelligence toward our goals, to link it to the aims we seek to achieve.

That is why I believe in the vital importance of this panel's work in this sphere, in the task of going where regulation cannot, to attain the best possible artificial intelligence. As shown in the video, we are rolling out specific programmes for artificial intelligence to improve people's lives in multiple areas.

In short, this is Spain's ambition in this field-not to participate in the artificial intelligence race, but to do our part, humble as it may be, in building artificial intelligence that is responsible and human-centred, that contributes to the well-being of all, leaving no one behind.

I would like to conclude with a topical development that I think is very apposite. Apollo Research and OpenAI recently published a study on the reasoning employed by the most advanced artificial intelligence systems and discovered something disturbing. In their internal reasoning, this intelligence, these models, use a specific term for the humans who monitor them. The word is "watchers". What is most revealing is not just this term, but the fact that these models act differently when they are aware of being observed.

This panel, this meeting, is our response to that approach. We are not going to be "watchers". We are not going to limit ourselves to noting that the most powerful artificial intelligence in history decides for itself when to follow the rules and when to ignore them. We must be leading actors in cooperation, maximizing the multilateral tools at our disposal to defend the common good and to champion all the benefits of a technology that is, without question, going to change the world's destiny.

It is up to us to ensure that this change is for the good, and for the betterment of humankind. So I wish you the best of luck and bon courage with this huge undertaking and responsibility that you have ahead of you, because the future of our species is at stake. No pressure. Thank you very much.

(Transcript edited by the State Secretariat for Communication)