EOSC Symposium 2025: Another Milestone for the EOSC Federation
In early November, Brussels hosted the EOSC Symposium 2025, one of Europe’s key annual events dedicated to the future of open science and research data management.
Artificial intelligence should carry Europe’s ethical DNA, experts said in Copenhagen. Is it realistic?
“Welcome to the real world,” opened the moderator of the AI in Science Summit 2025, held in Copenhagen on 3–4 November 2025 as a fully in-person event. “I hope your bodies remember that you were actually here,” he joked, setting the tone for a discussion that blended politics, science, and moral dilemmas — at a time when Europe is searching for its own path in artificial intelligence. Despite a packed agenda, the summit provided enough space for networking and even a screening of the movie The Best Option.
Europe seeks both inspiration and identity
Serge Belongie, Professor of Computer Science at the University of Copenhagen and a leading figure in the European Laboratory for Learning and Intelligent Systems (ELLIS) network, noted that Denmark’s ambitions reach beyond its current EU Council Presidency. Both ELLIS and the European Molecular Biology Laboratory (EMBL) can be viewed as an inspiration of how long-term funding can sustain excellent research.
Rebecca Adler-Nissen, Director of the National Center for AI in Society (CAISA) in Denmark, reminded the audience that artificial intelligence is not primarily a technical challenge, but a societal one. Her key question: How can AI strengthen democracy?
Democracy, trust, and life in a data-driven age were recurring themes. According to the speakers, it is essential to protect the individual experience of choice and meaningfulness, rather than replacing it with algorithms. Trust, they argued, should be created by people, scientists and teachers, not technology, and Europe should develop an immune system against distrust.
Fear of the snake bite
David Dreyer Lassen, Rector of the University of Copenhagen, expanded on this metaphor by recalling a scene from the film Oppenheimer, in which J. Robert Oppenheimer and Niels Bohr discuss the danger of lifting a stone without being ready for the snake beneath it. Bohr was referring to the immense power of nuclear energy and its potential peril to humanity. Today, David Dreyer Lassen suggested, Europe faces a similar situation with artificial intelligence, as the field remains dominated by the United States and China.
Denmark’s Minister for Higher Education and Science, Christina Egelund, echoed Marie Curie-Skłodowska’s famous words: “Now is the time to understand more, so that we may fear less.” According to her, AI will never replace human curiosity and it should carry Europe’s ethical DNA.
Artificial intelligence hallucinates to please people
The strongest warning came from Yoshua Bengio of the University of Montreal, who pointed out that less than one percent of global AI research focuses on safety.
Current models, he claimed, often “hallucinate” truths to please users which is something unacceptable in science. Designing “honest AI” is possible, but we must act quickly. Advanced systems already show signs of avoiding corrections or manipulating their trainers - behaviour that could lead to sabotage or data leaks. Accordin to Bengio the smarter AI gets, the better it deceives. Humans do this naturally but do we really want machines to imitate us that closely?
His vision is an AI that serves as intelligent scientific models, free from human biases and explains data rather than imitating people.
When asked how Europe could achieve this, he replied that instead of maintaining 27 separate systems, it would be better to have perhaps three unified ones, to secure Europe’s future in the global AI race.
Artificial intelligence and European values
Executive Vice-President Henna Virkkunen agreed that Europe needs AI that is competitive, human-centred, and trustworthy. Together with Commissioner Ekaterina Zaharieva, she pressed a red button on the podium to launch the pilot phase of RAISE – Resource for AI Science in Europe. This virtual institute aims to bring together essential resources for developing and applying AI and will be funded with €107 million from the Horizon Europe programme.
At the close of the summit, participants had the opportunity to watch the European Parliament’s sci-fi film The Best Option (2022), which vividly envisioned the future of artificial intelligence. In a way anticipating how today’s AI models actually operate. The screening served as a reminder that the debate on AI is not only political and technical, but also deeply cultural and existential.
