Superintelligence Europe — No. 010

Mistral AI announced Monday that it has secured $830 million in debt financing — its first debt raise in three years of operation — to fund 13,800 Nvidia GB300 GPUs for a new data centre at Bruyères-le-Châtel, south of Paris, operated by Eclairion. The facility will deliver 44 megawatts from launch and is scheduled to be operational by the end of June 2026. Seven banks arranged the financing: Bpifrance, BNP Paribas, Crédit Agricole CIB, HSBC, La Banque Postale, MUFG, and Natixis CIB. Total Mistral funding reaches $2.9 billion — the most capitalised European LLM developer.

The strategic shift is fundamental. Until now Mistral relied on Microsoft Azure, Google Cloud, and CoreWeave to run its models and serve customers. The Bruyères-le-Châtel facility is Mistral’s first owned compute infrastructure — the moment cloud dependency becomes infrastructure ownership. Mistral’s ARR crossed $400 million in February 2026, up from $20 million a year earlier, with a $1 billion target by year end. Clients include the French military and European governments seeking AI independent of US cloud providers. The Paris data centre is the near-term deployment. Running separately: a 1.4 gigawatt AI campus with MGX, Bpifrance, and Nvidia, construction beginning H2 2026. Plus a €1.2 billion commitment in Sweden. Mistral’s target: 200 megawatts of European compute by end of 2027.

A landmark paper combining international worker and firm survey data from 2025 and 2026 — co-authored by economists at the St. Louis Fed, Harvard, Vanderbilt, and WZB Berlin — was widely covered on Monday, providing the most rigorous academic measurement of the Europe–US AI adoption gap to date. The headline: 43 percent of US workers use AI for their job, versus 32 percent of European workers. At firm level: 7 percent of US companies use AI for production, versus 4 percent of EU companies.

The variation inside Europe is as important as the headline gap. In the six-country worker survey, AI use ranges from 36 percent in the United Kingdom to 26 percent in Italy. Across all 27 EU member states at the firm level, Sweden leads at 9 percent of firms using AI for production; Serbia sits at 1 percent. The paper finds this variation correlates most strongly with whether firms actively encourage AI use by employees — not with sector, education, or job type. AI-related time savings already represent 2.3 percent of total working hours in the US versus 1.4 percent in Europe. No evidence of systematic job displacement was found in either region.

Espoo-based IQM Quantum Computers announced Monday that it has secured a €50 million financing package from BlackRock-managed funds. The capital will accelerate IQM’s technology roadmap, expand R&D, and support entry into new markets ahead of the company’s planned public listing — through a SPAC merger with Real Asset Acquisition Corp — expected to close around June 2026 at an implied pre-money equity valuation of approximately $1.8 billion. A dual listing on the Helsinki Stock Exchange is under consideration alongside the US primary listing. The financing was secured before the SPAC announcement and is designed to lower IQM’s cost of capital as it scales toward commercialisation. Total funding to date exceeds €600 million.

Founded in 2018 as a spin-out from Aalto University and VTT, IQM builds full-stack superconducting quantum computers and sells on-premises deployments to research institutions, universities, high-performance computing centres, and national laboratories — giving customers direct ownership and control of their quantum infrastructure rather than cloud dependency. The company recently deployed the Aalto Q20 system, its fourth Finnish deployment. That BlackRock — the world’s largest asset manager — is now backing a Finnish quantum spin-out ahead of a US listing is a signal that institutional capital is moving beyond early-stage quantum research toward companies that can industrialise delivery at scale.

Stellantis — the Franco-Italian automotive group behind Fiat, Maserati, Peugeot, Jeep, and eleven other brands — announced Monday a five-year expansion of its partnership with Palantir Technologies, centred on deeper deployment of Palantir’s Foundry data platform and Artificial Intelligence Platform across global operations. The expanded agreement moves AI from isolated pilots into core business functions: supply chain decision-making, manufacturing quality management, demand forecasting, and data governance across brands and major production regions in Italy, France, Germany, Poland, and Spain.

The practical step-change from the original partnership to the renewal is the move from analytical insight to agentic execution. Rather than simply surfacing data for a human to act on, Palantir’s AIP is now being deployed to autonomously re-route supply chains when disruptions are detected, adjust production schedules in response to demand signals, and flag quality issues in real time across the manufacturing floor. Stellantis is separately targeting €20 billion in AI-driven software revenue through its SmartCockpit vehicle platform by 2030. The Palantir expansion is the operational AI infrastructure that makes that revenue ambition credible.

Credible accounts and publications that drove Monday’s European AI conversation. Filtered for genuine signal.