Artificial Intelligence Action Summit Press Kit

PRESS KIT 6-11 FEBRUARY ARTIFICIAL INTELLIGENCE ACTION SUMMIT

CONTENTS Editorial by the President of the French Republic P. 4 Introduction by the Special Envoy of the President of the Republic P. 6 France: a global leader in artificial intelligence P. 8 An Action Summit P. 12 Five key work tracks P. 18 A unique, inclusive approach P. 23 Programme P. 24 Press information P. 25

4 Editorial by the President of the French Republic The development of artificial intelligence is con- stantly gaining pace, and its immense potential is transforming our lives. It can predict, calculate and reproduce; it can accelerate the progress of our knowledge, our technology and our creativity. This technology brings many great opportunities. Everyone can see how the fields of health, edu- cation, work and innovation will be transformed forever by technologies with a potential that has not yet been achieved. Their deployment also brings risks and legitimate concerns, as regards the reliability of information, the protection of rights, and its accessibility to everyone. It is time to act, to preserve our heritage of thought and the balance of our societies, and to shape artifi- cial intelligence in accordance with the universal values we uphold. During the European Renaissance, the growth of science liberated men and women, reinvented medicine, brought political and religious changes, and transformed humankinds vision of nature and the universe. In the midst of this effervescence, a certain idea of humanity emerged: machines were subordinate to the mind, the universe to reason, and the individual to society. Similarly, as artificial intelligence emerges, I believe that France and Europe must ensure that these global shifts bear the mark of humanism. The future of artificial intelligence is a political issue, one of sovereignty and strategic independ- ence. To ensure our continent and our country, which has a rich and vibrant AI ecosystem, are leaders during this century, we need to support our talents and our businesses.

5 To continue innovating and creating, we need now to identify and anticipate the energy issues around large-scale deployment of artificial intel- ligence and the challenges for the world of work and creativity, and to enable every citizen, every company and every country to access reliable artificial intelligence. Building trusted artificial intelligence systems, following on from the Bletchley Park and Seoul Summits, sharing benefits on a vast scale, guaran- teeing independent access to artificial intelligence resources, and defining an inclusive and effective international governance framework, are the pri- orities of the AI Action Summit. This Summit is focused on action. To succeed, we will need the commitment of all stakeholders, including major companies, start-ups, researchers, students, States and NGOs. We can and must build artificial intelligence systems that serve everyone, that are compatible with our world view, that are steered through inclusive governance, and that preserve our sovereignty. Everybody will have a role to play, as Paris will show from 6 to 11 February.

6 Introduction by Anne Bouverot, Special Envoy of the President of the Republic Artificial intelligence has been arousing unprec- edented interest for months. For example, six months after the 2024 Olympic Games, Paris, France and Europe will be the centre of the worlds attention during the AI Action Summit (6-11 February 2025), which will focus on the future of our societies. These are set to be transformed by the growing development of artificial intelligence and its uses. France has been thinking about this since 2018, when it published the first part of its national strat- egy on the subject, and the National Commission on Artificial Intelligence, which I was honoured to co-chair, has worked on the issue more recently. Internationally, France is one of the seven coun- tries that take part in all of the main coordination mechanisms on artificial intelligence, which are multiplying. The AI Action Summit is the fruit of long and demanding work, building on previous events and marking the start of new momentum for international cooperation. Firstly, this Summit is unprecedented in its inclusiveness. Inclusiveness has been the watchword in prepar- ing this event. We have closely involved more than 700 stakeholders within our five working groups and consulted 800 companies and civil society organizations in all areas. Across the world, respondents have said they do not want to be passive consumers of artificial intelligence services and want to be included in their design and governance. Inclusiveness is also about access for everyone. We want to demystify artificial intelligence and its complexity, in a shift from science fiction to reality. We will spend a week showing uses in all fields, highlighting artistic creativity and involving the worlds young people.

7 Moreover, this Summit will be a week of tangible action. It is time for us to ensure, together, that the public interest is central to the development of artificial intelligence. In Paris, promotion of public interest AI will include the showcasing of 50 innovative projects addressing the sustainable development goals (SDGs). It will also include the launch of a large-scale initiative aimed at making access to artificial intelligence more equal across the world. This Summit will be an opportunity to get to grips with essential issues, including the quality of public and democratic debate and access to information in the generative AI age. Lastly, this Summit is a call to take a long view: firstly, through longer-term initiatives, which we will promote energetically from the Summit onwards; and secondly, the absolute need to conceive the two great transitions of our time the digital and environmental transitions together for the first time on the international stage, to make artificial intelligence a driver of progress, environmental protection and the public interest.

8 France: a global leader in artificial intelligence Building on the ambitious strategy driven by the French President since 2017, France is now positioning itself as a global leader in artificial intelligence. AN AMBITIOUS STRATEGY DEPLOYED SINCE 2018 Since 2018, France has been implementing an artificial intelligence strategy built around research, training and wide adoption of artificial intelligence by economic stakeholders. Some 2.5 billion has been deployed under the strategy to date, helping to bring about a leading global AI ecosystem . France in the artificial intelligence sector: More than 1,000 artificial intelligence start-ups that have raised money in France, including 1.9 billion in 2024; The leading global destination for foreign investments in artificial intelligence; The top host for the AI research hubs and decision-making centres of global leaders, such as OpenAI, Alphabet (Google), Tata Sons, Cisco, Criteo, DeepMind, Fujitsu, HPE, IBM, Intel, Meta, Microsoft, NaverLabs, Samsung, SAP, and Uber; Some of the worlds top champions, including Mistral AI, H, Hugging Face, PhotoRoom and Aqemia; 16 unicorns, including ContentSquare, EcoVadis, ManoMano, Meero, Mirakl, Payfit, Qonto and now Mistral AI; Some of Europes most powerful computing centres, both public (Jean Zay GENCI) and private (Scaleway, Outscale and OVH); Training and development of some of the worlds most powerful foundation models , including Llama 2, Llama 3 and Mistral Large. One of the greatest successes of the French artificial intelligence strategy has been to posi- tion France as a key actor in research. The strat- egy draws on excellent training, development of public and private computing power, and deploy- ment in all economic sectors. When it comes to training, the creation of four interdisciplinary artificial intelligence insti- tutes (3IA) has boosted French research by strengthening ties between public and private research and establishing 40 research chairs and 300 funded doctorate contracts; In computing power, the Jean Zay supercom- puter was created in 2019 and has supported more than 1,200 academic and industrial projects; Sectoral initiatives have helped structure and exploit data to transform public sector action in the health and labour fields, while defence and security projects have enjoyed significant resources.

9 These efforts, which received considerable support under the France 2030 plan, focused on four priorities: Doubling the number of AI talents by con- solidating top-quality research and training: 9 international AI Clusters have been created, with 360 million in funding. The 9 selected AI Clusters are: PR[AI]RIE PSAI (Universit PSL); MIAI Cluster (Universit Grenoble Alpes); Hi! PARIS Cluster 2030 (Institut Polytechnique de Paris); PostGenAI@Paris (Sorbonne University); ENACT (Universit de Lorraine); DATAIA-Cluster (Universit Paris-Saclay); ANITI (University of Toulouse); EFELIA (3IA Cte dAzur Universit Cte dAzur); SequoIA (University of Rennes). These investments have considerably strength- ened the artificial intelligence research and train- ing ecosystem across France. They should help combine this training, research and innovation quality to foster the emergence of academic leaders in France. The aim for 2030 is to train 100,000 people, including 20,000 in continuing training , and to ensure at least one of our centres of excellence features in the leading international rankings. In computing, multiply public and private investment. In the public sphere, Jean Zay has also been expanded and gradually opened up to economic stakeholders, and a European investment is planned to set up an exascale computer in Europe. Priority research programmes and infrastructures (PEPR) for artificial intelligence have been launched to work on frugal and trusted AI. In the private sector , several major projects have been launched in France, including Microsofts announcement at the 2024 Choose France Summit (4 billion), and more are to come. Support for the development of start- ups to foster competitive French businesses, through several actions: 1) development of technological building blocks for embedded artificial intelligence, and their integration within complex systems; 2) development of trusted AI systems, with the construction of the confiance.ai programme; 3) creation of digital commons throughout the generative AI value chain, to boost the development of the sector in France. Deployment of artificial intelligence throughout companies: align AI supply and demand and thus speed up the adoption of the technologies in companies, including through the AI Booster programme. Open source projects and the development of frugal artificial intelligence demonstrators supporting the ecological transition at local levels also help to ensure these technologies serve the many and the public good. In addition to becoming Europes leading gener- ative AI hub, France has risen from 13th place in the Global AI Index in 2023 to 5th place in 2024. Concerning the public-sector led dimension of the strategy (research, training and infrastructure), France has risen from 5th place globally in the 2021 Global AI Vibrancy Ranking to 3Rd in 2023..

10 FRANCE BOASTS EXCEPTIONAL ASSETS WHEN IT COMES TO ARTIFICIAL INTELLIGENCE Its excellent schools and universities enable it to train some of the worlds best engineers, who are often behind cutting-edge technologies. Businesses benefit from the scientific and engineering skills of talents trained in France , who have an international reputation. For exam- ple, France maintained its 7 th place position in the global ranking, and second in the European ranking, for publications at scientific conferences on artificial intelligence. France also stands out for its unique open source ecosystem, with emblematic projects like Mistral AI, Hugging Face and scikit-learn that are some of the worlds leading open source AI libraries. These initiatives are supported by leading public research centres such as the National Institute for Research in Digital Science and Technology (INRIA), National Centre for Scientific Research (CNRS) and Alternative Energies and Atomic Energy Commission (CEA). Moreover, attractive tax schemes such as the research tax credit (CIR), the young innovative company (JEI) status and the French Tech Visa strengthen this dynamic ecosystem. Since 2018, many companies have chosen France to establish or expand their artificial intelligence laboratories, including OpenAI, Alphabet (Google), Tata Sons, Cisco, Criteo, DeepMind, Fujitsu, HPE, IBM, Intel, Meta, Microsoft, NaverLabs, Samsung, SAP, and Uber. Meanwhile, France had 590 arti- ficial intelligence start-ups in 2023 (up from 502 in 2021), which raised more than 3.2 billion in 2023 six times more than in 2018. Above all, France offers abundant, flexible and decarbonized energy at competitive prices thanks to its mainly nuclear power sector.

11 FRANCE PRODUCES ABUNDANT, COMPETITIVELY PRICED AND DECARBONIZED ELECTRICITY, FOR ARTIFICIAL INTELLIGENCE THAT PROTECTS THE CLIMATE. Some 93% of the electricity produced in France was decarbonized in 2023. In his 2022 Belfort speech, the President set a clear course to consol- idate Frances leadership in the energy transition, with two priorities: massive development of renew- able energy sources and new momentum for the nuclear sector. Today, this priority is clear in the objectives to deploy 100 GW of solar capacity and 40 GW of offshore wind capacity by 2050, when it comes to renewables, and in the announcement of six new nuclear reactors. These investments will ensure France remains a massive producer of decarbonized energy. Having exported a net 89 TWh in 2024, France beat its own historic record, helping its European neighbours to achieve our Paris Agreement climate goals. With its very competitive current nuclear power infrastructure, massive investment in solar and wind power, and a very large and robust high and very high voltage power grid which enjoys constant investment, France offers very condu- cive economic conditions for projects requir- ing large quantities of clean electricity, such as data centres. In the context of the reform of the European electricity market and the regulatory framework replacing the French mechanism for regulated access to historic nuclear energy capac- ity (ARENH), the French Government has made the reforms needed to guarantee competitive and stable prices for businesses, protecting them from volatility linked to oil and gas prices. Several types of contract are available to countries requiring a lot of electricity, helping guarantee stable and competitive prices for five to fifteen years. This modern energy policy, seen by many actors as being at the forefront when it comes to address- ing climate and competitiveness challenges, as well as Frances good geographical position at the heart of Europe, makes the country a strategic hub for data centres, boosted by major invest- ment in supercomputers like Jean Zay and the future Alice Recoque exascale computer. Private companies like OVHcloud and Scaleway, as well as international actors like Microsoft, Amazon and Telehouse, have also invested massively in France. The bill on simplifying the business environment that will be soon be tabled before the French par- liament includes a measure to classify data cen- tres as projects of major national interest (PINM), enabling expedited administrative procedures. Government-secured PINMs enjoy faster author- ization procedures for planning and access to the electricity grid. Thanks to exceptional talents, a vibrant ecosys- tem and an ethical vision, France has earned its place as a global leader in artificial intelligence. Driven by its ambitious strategy and structural investments, it is in a position to rival the great AI nations while deploying its own vision for the technologies.

12 An Action Summit THE MAIN FOCUSES OF THE SUMMIT Artificial intelligence (AI) is more than just an industrial and technological revolution . It has the potential to bring about a profound para- digm shift in our societies, in how we relate to knowledge, work, information, culture and even language. This technological revolution knows no borders. In this sense, artificial intelligence is a scientific, economic, cultural, political and civic issue requiring intense international dialogue involving the governments, researchers, busi- nesses, creative professionals and civil society, to ensure the science, solutions and standards that shape artificial intelligence of the society we want to build for tomorrow are developed collaboratively . And so France, at the initiative of its President, has decided to build on the momentum generated by the United Kingdom and the Republic of Korea, which organized the previous two Summits, by offering to host the AI Action Summit in Paris on 10 and 11 February 2025 . The Artificial Intelligence Summit will enable a dialogue with a large number of countries on artificial intel- ligence, including the United States, China, the major emerging countries and many partners from every continent, including Africa . It will be co-chaired with India, which is fully involved in its preparation, as a member of the steering commit- tee and the five working groups, and is co-chairing the working group on international working group on the international governance of AI. This international Summit will bring together almost 100 countries and over a thousand pri- vate sector and civil society representatives from across the world, invited on an inclusive basis in recognition of their commitment to the action promoted by the Summit and their desire to con- tribute actively to debate. The question we all face as the worlds citizens and users, start-ups and major corporations, researchers and decision-makers, artists and media outlets is ultimately a simple one: How can we massively develop artificial intelligence technologies and uses across all world countries? How can we ensure nobody is left behind and preserve our freedoms in the AI revolution? How can we ensure that uses of artificial intelligence respect our humanist values and that the technology serves society and the public interest? The stakes are sky-high: we must enable artificial intelligence to fulfil its initial promise of progress and empowerment in a context of shared trust that contains the risks inherent to technological development, while seizing every opportunity .

13 It is in this context that France and Europe are hosting this Summit, in line with our clear commitment to open and inclusive progress on artificial intelligence and to an approach focused on sovereignty and strategic inde- pendence when it comes to absolutely crucial technologies for our societies and economies . The French President has been working to make France and Europe leading actors in arti- ficial intelligence since 2017 , as illustrated by how the national strategy is fully integrated into support for the development of the technologies at European level. Europe is taking a unique, cre- ative and responsible approach to digital tech- nology, combining innovation and regulation to serve its citizens. It is also an attractive place to work for artificial intelligence researchers and developers . Moreover, Europe has a singu- lar message when it comes to addressing major global challenges, which require an inclusive approach like the sharing of vaccines during the pandemic.

14 AIMS OF THE SUMMIT The AI Action Summit will be an opportunity to commit together to develop the science, solutions and standards that will ensure that artificial intelligence serves the fundamental public interest . Artificial intelligence has incredible potential and could resolve a great many very complex prob- lems , seeking new therapies, predicting climate hazards and discovering new scientific correla- tions through analysis of large datasets. At the right scale and with the right governance, artificial intelligence can help develop, for example, tailored and effective education models that support both learners and teachers. The use of artificial intelli- gence can be advanced in many sectors, including health, science and, more generally, fields where it can boost productivity considerably. The AI Action Summit and its accompanying AI Action Week should highlight these innovations and their development in France, in Europe and in all participating countries, so that everyone can get an idea of the impact and the potential of AI. It is an important time to develop the ecosys- tems that allow the development of AI and its swift roll-out, from infrastructures to uses and models . All actors in the value chain have therefore been invited and their role will be to demonstrate and share best practices and the technical, economic or social challenges of their technol- ogies . Countries will be invited to present their measures and practices that foster the devel- opment of artificial intelligence and their swift roll-out. The Summit should also enable progress to: 1) Open up access to independent, safe and reliable artificial intelligence for the many , so that everyone, whatever their origin and knowl- edge of the technology, can adopt and benefit from it. In order to reduce the growing digital divide in the context of excessive concentra- tion in the artificial intelligence market, in the hands of a few private actors , this Summit will also seek to demonstrate the tangible, everyday benefits that it can have for all the Sustainable Development Goals, based on projects worked on already by both private companies and govern- ments across the world. The Summit should also foster a large-scale ini- tiative for public interest artificial intelligence , to create a more diverse, open and accessible ecosystem for everyone. That should mean very tangible support both for development but also for the sharing of key blocks, artificial intelligence public goods, such as structured and accessible datasets, open source tools and the training of tomorrows talents. In order to ensure sufficiently broad reach, this project should be supported by both the public and private sectors and benefit everyone to serve the public interest. In particular, this means we can respond to the strong demand from our partners in emerging and developing countries, who would like to have access to arti- ficial intelligence services and boost their skills in the field, while ensuring their strategic inde- pendence and sovereignty . 2) Develop more frugal, environmentally friendly artificial intelligence systems to address the climate and energy transition. While artificial intelligence can help fight global warm- ing and protect ecosystems, it is currently on an untenable trajectory when it comes to energy use, particularly in terms of fossil energy con- sumption . The latest forecasts suggest that the energy needs of the artificial intelligence sector will be ten times higher in 2026 than in 2023.

15 That is not sustainable. In response, the Summit must enable all types of stakeholders, from all countries, to commit to more sustainable arti- ficial intelligence . In real terms, that means commitments to use low-carbon energy, in-depth research into the environmental impact of arti- ficial intelligences development, evaluation of models on that basis, new standards, and greater green investment and investment in low-carbon energy sources at every link in the value chain. The Summit should also be a platform for those working to ensure the development and adoption of artificial intelligence happen in a social frame- work that respects our values, workers and cre- ative professionals . 3) Ensure global governance of artificial intel- ligence is effective and inclusive. All interna- tional leaders should be able to discuss all the concerns raised by AI : not only essential issues of ethics and security, but also those of protecting fundamental freedoms, fighting disinformation, safeguarding intellectual property, fighting market concentration and ensuring access to data. In September, the United Nations sketched out a vision for this global governance of artificial intel- ligence in a Global Digital Compact. We now want to help implement its recommendations . Private stakeholders and civil society actors must be included in this reflection, so that together an international AI governance structure can be implemented that combines the many challenges linked to this technology.

16 CONTRIBUTORS TO THE SUCCESS OF THIS SUMMIT A dedicated team has been working for almost a year to ensure the Summits success . The Special Envoy of the President of the Republic, Anne Bouverot, and five thematic envoys (Sana de Courcelles, Guillaume Poupard, Martin Tisn, Roxanne Varza and Henri Verdier) have been tasked with coordinating the work of contact groups on the Summits work tracks: Future of Work, Trust in AI, Public Interest AI, Innovation and Culture, and Global Governance of IA. A Secretariat has been set up to organize this inter- national meeting and its communication. Lastly, under the coordination of the Ministry for Europe and Foreign Affairs, a number of French minis- tries and public actors have contributed actively to its content and programme. France is not going to the Summit alone India will be a co-chair . Furthermore, at international level, the Summit team has been supported by a steering committee of representatives from some 40 countries and international institu- tions , as well as representatives of academia, the private sector and civil society. The Steering Committee has met five times. Its aim is to pro- vide high-level guidelines to the five groups led by the thematic envoys, which are responsible for jointly building the Summits outcomes. More than 700 participants (public and pri- vate sector partners, researchers, NGOs from around the world) have taken part in the five contact groups, meeting regularly from summer 2024. The membership of the Summits contact groups is available on its website1. The Summits envoys have visited many coun- tries around the world to prepare the event and gather recommendations from various pub- lic, private and philanthropic stakeholders. Five civil society meetings have also been organized, attended by more than 500 organizations. As the impact of artificial intelligence is unprec- edented and multidimensional, open and mul- tistakeholder processes are needed to ensure todays decisions are informed and legitimate. Thousands of stakeholders have therefore been able to take part in preparing the Summit through various channels: By proposing an event that could be part of the Road to the Summit , a series of prepa- ratory events worldwide to discuss the issues addressed at the Summit. These events have involved public and private stakeholders, experts and academics. The cultural and media sectors have also taken part. The Road to the Summit ends on 11 February, with a series of Summit side events. The list of these events is on the Summit website2; By taking part in the Public Interest AI call for projects , organized by the Paris Peace Forum, through which some 50 projects have been selected from around the world; By contributing to the 35 AI Convergence challenges organized by the Secretariat- General for Investment, illustrating the vibrancy of academic and industrial ecosystems and the solid foundations for innovation in which they operate worldwide; 1 https://www.elysee.fr/en/sommet-pour-l-action-sur-l-ia/presentation 2 https://www.elysee.fr/en/sommet-pour-l-action-sur-l-ia/join-the-road-to-the-summit

17 By submitting written contributions and taking part in the citizens consultation orga- nized by the AI & Society Institute (ENS-PSL), the French National Digital Council (CNNum), The Future Society, Make.org and the Sciences Po Technology and Global Affairs Innovation Hub, and which generated several thousand contributions. By taking part in the international call for expressions of interest launched by France in the artificial intelligence sector, as part of the AI for Efficiency initiative highlighting tan- gible examples of successful use of artificial intelligence by businesses of all sizes, public administrations and international organiza- tions in order to generate momentum among companies, reassuring them so they can make the leap. THE EUROPEAN DIMENSION OF THE SUMMIT The European Union has been particularly active in the various preparatory forums throughout the preparation of this Summit. The European Commission and its Member States have contributed to this work within both the Steering Committee and the contact groups. The Summit has also been discussed within the Council of the European Union. The Summit will therefore reflect a balanced European approach to artificial intelligence that combines support for innovation, adequate regulation and respect for rights to ensure devel- opment of these technologies in the interests of all, including developing countries. Europe can and must significantly strengthen its positioning on AI and accelerate invest- ments in this field , so that we can be at the forefront on the matter. The Summit will be the opportunity to discuss these issues so that Europe can become an ideal place to train talents, set up infrastructures, develop models and roll-them out massively and invest in AI. The European dimension of the Summit is focused on essential deliverables for the development of a technology that Europe must maintain mastery of through (i) talents, research centres and cutting edge development, by establishing an assertive European attractiveness strategy; (ii) rigorous and compatible data management to foster the development of artificial intelligence; (iii) dedicated resources to strengthen expertise and foster emergence of global leaders; and (iv) a coalition of European companies that make the leap and prepare the groundwork to make Europe an incubator and a general hotbed for the growth of future innovative leading actors, which will become tomorrows international artificial intelli- gence champions. Lastly, the European Union is a direct stake- holder of the Summit thanks to the Digital Europe Programme, which has contributed to organizing the Summit on 10 and 11 February 2025.

18 Five key work tracks No subject will be avoided: from the future of work to frugal AI, from the safety of models to the specific issues of the creative professions, and from the need for linguistic and therefore cul- tural diversity to protection of privacy. INNOVATION AND CULTURE The innovation we are seeing in artificial intelligence is deployed very fast and affecting every sector. It can be very beneficial, so long as an appropriate framework is defined from the outset. That is particularly true when it comes to culture, a sphere with specificities where the issue of intellectual property is particularly sensitive. Roxanne Varza, Summit Envoy for Innovation and Culture. Artificial intelligence can offer significant pro- ductivity gains, enhancing organizations com- petitiveness and innovation. The dissemination of artificial intelligence should therefore be accelerated but also supported, by offering independent solutions and which ensure the sovereignty of the countries that use them . The Summit will be an opportunity to highlight uses and new technologies that can respond to unprecedented challenges. The Summit should also make it possible to iden- tify the practices and methods that enable the ecosystems of artificial intelligence actors to develop, particularly in France and Europe but also all around the world. The development of infrastructures and data- bases will also be a focus of the summit, as an essential driver of AI development. The use of artificial intelligence by all types of organizations and people to contribute to addressing certain major current challenges seems essential . Above and beyond the issue of the competitiveness of organizations and nations, artificial intelligence needs to be deployed appropriately , serving environmental protection and cultural diversity. The launch of an ambitious project to encourage public interest innovation is planned. This project will speed up the deployment of artificial intel- ligence in more and more languages, countries and sectors, while lowering barriers for smaller innovators and adopters . The importance of respecting intellectual prop- erty rights , particularly copyrighted content, and the compensation of content creation should also be reasserted. That will be a key focus of the Summit. Collective efforts to foster the production and distribution of reliable information produced by media outlets and journalists , and the sharing of tools to fight disinformation, should be welcomed. In this context, equitable business models need to be developed that benefit all stakeholders . Examples of cutting-edge artificial intelligence use and support for virtuous competition models will be showcased.

19 The Summit will be a unique opportunity to pro- mote a more sustainable model for artificial intelligence internationally . This is the result of collective work that iden- tified the needs and viewpoints of more than 150 organizations in 20 countries . This work involved governments, civil society organiza- tions and other participants. The excellent work by existing organizations and initiatives such as the International Energy Agency (IEA), the Organisation for Economic Co-operation and the Development (OECD) and the United Nations Educational, Scientific and Cultural Organization (UNESCO) has been a precious asset. FUTURE OF WORK Artificial intelligence can become an incredible ally if we join forces to make it a factor of social and economic progress, reconciling productivity and wellbeing at work, to benefit all workers. Sana de Courcelles, Summit Envoy for the Future of Work. The development of artificial intelligence is pro- foundly reshaping labour markets, jobs and work- ing methods. The Summits work aims to seize opportunities while containing risks and fostering working tools that enhance productivity, safety and wellbeing at work. As the AI revolution begins within organizations, its impacts must be apprehended immediately and must be tracked, and it is important to foster the sharing of knowledge and best practices. The ambition of the members involved in the Summits Future of Work track is to promote socially responsible use of artificial intelligence through sustained social dialogue . To share this information beyond the Summit itself, work has begun on an initiative to build measurement capabilities regarding the impact of artificial intel- ligence and algorithmic management of labour. The Future of Work team is working actively with its partners to create an effective network of observatories by connecting several inter- national, national and private sector bodies to improve sharing of knowledge. The challenge is also to encourage economic and social stakeholders to make economic and social commitments to sustainable and responsible arti- ficial intelligence at the Summit, involving almost 40 representatives of governments from around the world, international social partners, academia and experts, economic actors and businesses. Discussions held in the framework of the Summit should therefore supply relevant tools to the actors that wish to take the path of responsible AI in the field of work, in particular tools that avoid gender-based discrimination in access to employment, as well as the rolling out of a human resources policy that encourages training and adaptation to the new needs created by AI.

20 TRUSTED AI In light of the concerns aroused by the swift development of artificial intelligence, it is important to understand and apprehend risks, building on science and expertise to offer effective, accessible solutions that benefit everyone and protect our democracies. Guillaume Poupard, Summit Envoy for Trusted AI. Following the Bletchley Park and Seoul Summits, the international community now needs to con- tinue its efforts comprehensively and inclusively in order to identify and address the risks of artifi- cial intelligence. Safety and security are central to trust , which is a prerequisite for achieving our overarching aim: public interest artificial intelligence. The approach of the AI Action Summit is focused on three types of public goods: sci- ence, solutions and standards . The Bletchley Park Summit sparked an effort to establish the first public good in this field: a solid scientific con- sensus on artificial intelligence. The time has come to develop open, universally accessible technical solutions and work on common standards to avoid fragmentation and encourage convergence around high levels of safety and security for artificial intelligence. The risks associated with the development and integration of artificial intelligence systems will be addressed in Summit deliverables, particularly as regards the impact of AI on cyber security and on identifying and controlling the serious risks that could result from malicious use of these tech- nologies. Specifically regarding the challenges of AI in terms of information manipulation, several deliverables will assess the real impact of AI on the integrity of information, and also illustrate the opportunities and solutions offered by artificial intelligence in this area . The Summit will also be an opportunity to take stock of existing voluntary safety and security commitments on the part of major artificial intel- ligence pioneers. Issues around ethics and the fight against dis- crimination will also be discussed. These work focuses will be the subject of in-depth and sustained discussion between more than 20 States, the European Commission, the major international organizations working on these issues, some 50 civil society organizations and around 30 companies of all sizes from the sector.

21 PUBLIC INTEREST AI The choices inspired by artificial intelligence models will be crucial for our societies. They can guarantee that more or fewer women find work, influence the discovery of treatments for incurable diseases and improve or worsen equality of opportunity. These choices are collective choices. To realize the potential of artificial intelligence, society must remain in control. Martin Tisn, Summit Envoy for Public Interest AI. The current trajectory of artificial intelligence development will result in three major issues: Increased inequality between those who control and those who use artificial intelligence; Progress made in AI concentrated in a small circle of private actors, jeopardizing both the diversity of actors involved but also the sove- reignty of countries that do not have any leve- rage in this critical technology; Missed opportunities to resolve key social problems (such as fighting cancer) because of the fragmentation of public interest artificial intelligence initiatives and scarce data. That is why France wants, in the framework of the Summit, to see the creation of a new global plat- form that can serve as an incubator for artificial Intelligence serving the public interest, independ- ent solutions, open access or controlled access depending on resources, and which ensures that users have sovereign use of the technology. This is a collective effort, involving consultations with dozens of countries, hundreds of civil soci- ety organizations and businesses from across the world, aimed at developing artificial intelligence common goods in the field of data, open models and citizen participation that can be reused by all States and organizations that so wish, in order to address the aspirations shared by the stake- holders consulted. Since June, a contact group of more than 200 people has met three times and will continue to meet as this initiative takes shape. GLOBAL GOVERNANCE OF IA For the AI revolution to mean progress for everyone, it is essential to design a collective and inclusive governance architecture to address all the challenges of the technology. Henri Verdier, Summit Envoy for Global Governance of IA. The global governance framework for artificial intelligence is currently piecemeal . Some coun- tries and regional organizations, like the European Union, have established their own regulatory frameworks autonomously. At international level, however, there is no unified governance, even if many actors are and must be involved, as arti- ficial intelligence raises all sorts of questions , including safety, ethics, environmental impact, access, competition, cultural diversity, data pro- tection, interoperability of standards, military use, integrity of information, open technical standards and actors sovereignty.

22 However, in the absence of coordination, multi- ple parallel initiatives have been launched in very different formats, like the G7 Hiroshima AI Process and the Bletchley Park Summit in 2023, and the Tech Accord to Combat Deceptive Use of AI in 2024 Elections that was signed by compa- nies alone in Munich in February 2024. This creates a complex landscape that is often insufficiently inclusive and where frameworks sometimes compete. In this context, the account- ability of artificial intelligence actors is upheld very strictly in terms of the commitments demanded of them and at the same time very loosely due to the lack of competent control institutions. Meanwhile, various norms co-exist depending on the regions concerned, which penalizes smaller actors that are less well-equipped to navigate between legal systems. Echoing the vision for more inclusive and effective governance of artificial intelligence promoted by the Global Digital Compact, which now needs to be established tangibly, the primary aim of the Summit in this area is to clarify and design a shared and effective gov- ernance framework with all relevant actors so that science, solutions and standards are bet- ter shared. Inclusiveness is the watchword . Today, only seven countries worldwide are participating in the major international artificial intelligence ini- tiatives, while 119 are totally absent from them. Moreover, private actors and civil society need to be fully included in this governance, which must not be restricted to public authorities alone. As such, the chosen approach is a collective one , involving many consultations with various stakeholders from across the world. The aim is clear: to build a consensus on the global gov- ernance framework for artificial intelligence, with and for all parties . With 70 partners, including States, international organizations, businesses and civil society rep- resentatives, the aim is to survey the artificial intelligence governance system s needs, actors, fundamental principles, limits and room for progress. One single governance initiative is not the answer . Rather, existing initiatives, like the Global Partnership on Artificial Intelligence (GPAI) that will meet in an expanded format on on 9 February with the Minister for Europe and Foreign Affairs, Jean-Nol Barrot, to agree on a new action plan for 2025, need to be coordinated to build a global, multi-stakeholder consensus around an inclusive and effective governance system for artificial intelligence.

23 A unique, inclusive approach

24 Programme: an AI Action Week, from 6 to 11 February 2025 Events in relation to the Summit started several months ago, with the Road to the Summit . These events (around 100) around the world brought together hundreds of participants to discuss the Summits themes. In February, these international efforts will come to fruition in an AI action week that will start in Paris on 6 February and culminate with the Summit. These discussions will begin with science and presentation of the latest known and universally recognized academic elements on artificial intel- ligence. For example, the Institut Polytechnique de Paris (IP Paris) will host an international sci- entific conference on AI, Science and Society, on 6 and 7 February on its campus. addressing the transformations brought by artificial intelli- gence for science and societies. By fostering an interdisciplinary dialogue, this meeting will see presentations by leading researchers, including Nobel laureates, of their work, followed by various discussions between leaders in Paris throughout the week. 8 and 9 February will focus on culture and AI , with dedicated events in Paris, that will be open to the public . The aim will be both to present works produced by artists using artificial intel- ligence and to address the debates sparked by the emergence of AI technologies in this sector (copyright, heritage, shifts for museums, artistic potential expressed thanks to and using artificial intelligence, etc.). A dedicated press kit will be provided. The Summit will begin in the Grand Palais on 10 February , starting with a forum bringing together many stakeholders from around the world (including representatives of governments, businesses and civil society, researchers, artists and journalists). This first day of the Summit will involve conferences, round tables and presenta- tions, focused largely on solutions offered by artificial intelligence. It will enable many bilateral meetings between participants who are gathering for the first time in such a format, with particularly diverse stakeholders. At the end of the day, various evening side events will be organized by third parties. The French President will invite his counterparts and a few leading figures in artificial intelligence to a dinner at the lyse Palace . The Summit of Heads of State and Government will take place on 11 February , on the occasion of a plenary session at the Grand Palais gather- ing high-level guests from the private sector, civil society and international organizations to discuss among leaders the key common actions to take on artificial intelligence.

In parallel, the AI Action Week will culminate in a Business Day at Station F in Paris on 11 February , where businesses of all sizes, investors and financial institutions will hold discussions and illustrate tangible solutions deployed within organizations and to serve the public, report on their latest business and technology progress and talk about the global financial context for the sector. A number of side events will take place on the same day in Paris, including events dedicated to artificial intelligence and democracy and the envi- ronmental impact of these technologies, at the cole Normale Suprieure (ENS) and the Ministry for the Ecological Transition, respectively. At the same time, a crisis management exercise will bring together between 250 and 300 AI design- ers and producers, and cyber security experts, for a half day to study several cyber security crisis management scenarios involving AI systems. Press information SUMMIT DIGITAL COMMUNICATION VIA THE OFFICIAL WEBSITE AND SOCIAL MEDIA All information and content about the Summit will be posted on one institutional website in French and English: ai-action-summit.fr. Summit information will also be shared on the social media accounts of the French Ministry for Europe and Foreign Affairs: https://www.twitter.com/francediplo https://www.instagram.com/francediplo http://www.linkedin.com/company/ ministere-des-affaires-etrangeres-francais https://www.youtube.com/francediplotv flickr.com/photos/francediplomatie/albums ACCESSIBILITY Summit communication will adhere to the Charte daccessibilit de la communication de ltat , which contains all the rules and best practices regarding communication accessibility. This charter requires clear and easily understand- able information, taking into account the situa- tions and the needs of all people. It is available here. 25

Press contact [email protected]