
Fifty-nine of Europe's best-known AI scientists and company founders have written to the EU's 27 heads of state and government with one demand: fund AI properly in the 2028-2034 budget, or lose the ability to build the technology at home.
The open letter, titled "An Urgent Call: Europe Cannot Sleep Through the AI Revolution", went public on 18 November at the54.eu. It also addresses the presidents of the European Council and the European Commission. Its signatories come from every member state. They include economist Philippe Aghion, a 2025 Nobel laureate, AI pioneer Yann LeCun, Max Planck director Bernhard Schölkopf, and the chief executives of ElevenLabs, DeepL, Lovable, OVHcloud and InPost.
The timing is deliberate. EU leaders hold a special summit on the seven-year budget on 26-27 November, and the letter warns that the window for action is closing.
It also follows the informal European Council in Dublin, where three leaders made much the same argument in public. Poland's Donald Tusk called for pooled funding, leading AI research institutes and a working 28th regime for companies. France's Emmanuel Macron said European AI firms must find their computing power, their funding and their first major contracts in Europe. Sweden's Ulf Kristersson called for stronger European AI capabilities and fewer strategic dependencies.
The signatories now want those words written into the budget.
"We take the leaders at their word. What they said is good news for Europe, but what counts is delivery, not declarations. The test comes on 26 and 27 November, when the same leaders decide whether AI gets enough priority in the budget," said Piotr Sankowski, director of the IDEAS Research Institute in Warsaw, professor at the University of Warsaw and one of the signatories.
The current draft moves in the opposite direction. The Commission proposed €54.8bn over seven years for digital leadership, a line that covers AI, chips, cloud and cybersecurity together. The Council's October draft cuts that to €47.7bn, a 13% reduction, and reserves nothing for AI. Research funding falls from €175bn to €158bn.
The gap with the United States is wide. Amazon, Alphabet, Microsoft and Meta plan to invest about $730bn between them in 2026 alone, most of it in AI infrastructure. The EU hosts about 5% of the world's AI computing capacity, against 78% in the US, according to the Brussels think tank Bruegel. Europe and the US each create about 900 AI start-ups a year, Dealroom data show, but once those companies need to grow, European ones raised $12bn in 2025 against $141bn in the US.
These comparisons have limits. The $730bn is private company spending set against a public programme, and Bruegel finds that most planned European computing capacity is privately financed too. The 13% cut is also measured against the Commission's proposal: both the digital and the research lines still grow compared with the current 2021-2027 budget.
The letter sets out five priorities. It wants binding simplification of EU digital rules and a single legal regime that lets a company incorporate once and operate in every member state. It wants pension and insurance savings channelled into funds that back growing European companies, with governments acting as first major customers. It wants money set aside for AI in the European Competitiveness Fund and the next research programme. It wants jointly funded, independent AI research institutes on the model of the ELLIS Institutes. And it wants published annual targets for investment, computing power, talent and adoption, reviewed each year by heads of state and government.
The signatories are not neutral observers. Several run companies that would gain from public contracts and from more capital for European technology firms.
The letter ends on the line "What is measured at the top gets done", and two organisations plan to do the measuring. On 7 December in Brussels, the IDEAS Research Institute and the European Policy Innovation Council (EPIC), a Brussels think tank, will launch an AI Competitiveness Tracker. It will follow how far the EU and its member states have moved from AI strategy to AI adoption, benchmarked against the US and China.
"The ambition expressed in Dublin is welcome. But two years after Draghi, Europe still faces a regulatory paradox: leaders call for simpler rules and faster AI adoption, while businesses face overlapping laws that can pull in the opposite direction. EPIC's AI Competitiveness Tracker will independently assess whether Europe's investment and regulation support its ambitions—and whether we are closing the gap with our competitors," said Antonios Nestoras, founder and director of EPIC.
The arguments are familiar. Mario Draghi made most of them in his 2024 competitiveness report. What is new is the timing and the breadth of the signatures: one voice from science and business in all 27 countries, eight days before leaders divide up the next seven years of EU money.
AI has no natural defender in those talks. Farm and regional funds have national capitals fighting for them, while a digital line is easier to trim. The letter tries to change that by tying three leaders to their own public statements. The first test comes at the summit on 26-27 November. The second arrives in December, when the tracker starts recording what governments actually deliver.
Disclosure: EU Insider is published by the European Policy Innovation Council (EPIC), one of the two organisations launching the AI Competitiveness Tracker described in this article.
