The European Union has committed €18.4 million in direct funding to three new generative AI projects. This is part of a larger €36.8 million total investment. Outcomes are expected by 2029-2030. One initiative, the 'Floods and droughts' consortium, led by Italy, will receive €8.8 million. Its purpose: to develop Gen-AI tools for 'anticipation, crisis management and citizen communication' related to climate disasters. Public administrations across the bloc will develop these technological solutions for environmental crises.
The Digital Fortress
The 'Eunomia AI' consortium, led by Spain, will receive €6.9 million from the EU, part of its €13.9 million total value. This project aims to accelerate the adoption of 'trustworthy GenAI' in European public administrations. It involves 13 other EEA countries and plans to deploy AI in 25 public administrations and at least 15 public services by June 2029. Public administrations are the very bodies responsible for processing asylum claims, managing detention centres, and issuing deportation orders.
The project description states its goal to reinforce 'European tech sovereignty' by using and fine-tuning the EU's open-source models. This pursuit of technological independence for public services, which include those involved in migration management, aims to ensure EU control over the digital tools used within its administrative structures. A 'centralized, high-performance GenAI infrastructure' will be procured and hosted at the University of Granada, with an option for federated local nodes. Such centralised infrastructure could facilitate the sharing of data, including biometric information, across EU member states, further strengthening the surveillance capabilities of Fortress Europe.
AI for Exclusion, Not Aid
The third project, 'EuropAI,' led by the Netherlands, also receives €6.9 million in EU funding, part of its €13.9 million total cost. Its estimated completion date is June 2029. It aims to accelerate 'sovereign and trustworthy GenAI solutions' across European governments. This initiative focuses on creating a framework for member states to co-create, test, and share GenAI solutions in three strategic domains: 'legislative and administrative simplification,' 'urban analysis and spatial planning,' and 'digital assistance for citizens across web and mobile.' These domains directly intersect with the lives of migrants and asylum seekers, from the complexity of asylum laws to urban planning that often excludes or segregates migrant communities.
The consortium describes European public administrations as relying on 'imported 'AI fast food' from non-EU providers, like ChatGPT, Copilot or Gemini.' Their plan is to offer a trusted 'community kitchen where Member States co-create, test, and share GenAI solutions.' This system's backbone is built on Denmark's OS2ai platform, which already serves more than 28,000 employees, and the Netherlands' GovChat-NL. This push for EU-controlled AI systems can be seen as a move to ensure that the tools used in public services, including those that manage migration, are free from external oversight or influence, further entrenching the EU's border logic.
The Cost of Digital Walls
The EU faces a 'delicate balancing act,' according to the article, between encouraging AI adoption and avoiding 'AI-washing' projects that promise innovation but deliver little. Past EU-backed schemes have seen this outcome. This €36.8 million investment, with €18.4 million directly from the EU, is directed towards technological solutions for public administrations, with final outcomes expected by 2029-2030. The projects aim to reinforce 'European tech sovereignty' and improve 'crisis management' for climate disasters, alongside 'legislative and administrative simplification' and 'urban analysis.' No mention is made of how these advanced digital tools will facilitate safe passage or ensure the rights of those seeking asylum.