Seminal
The magazine and atlas of EU-funded researchA new publication from the web and AI team at Scott Charnley

Atlas / Spain

Marie Skłodowska-Curie Action · Running

AIStars4EU European Research Mobility Network for Sovereign, Scalable and Trustworthy AI Systems

Coordinator
Universitat Rovira I Virgili, Spain
Runs
1 January 2027 to 31 December 2030
Status
Running
EU funding
€1.6m
Total cost
—
Partners
22 organisations in 12 countries
Type
Marie Skłodowska-Curie Action
Grant agreement
101341806

What the project is doing

In the project's own words, from its summary on CORDIS.

AISTARS4EU aims to strengthen Europe’s technological sovereignty in Artificial Intelligence by advancing scalable, secure, and energy-efficient infrastructures for generative and agentic AI across the cloud–edge–HPC continuum. The project addresses a critical gap between cutting-edge AI capabilities and their reliable, sustainable deployment in real-world environments, aligning with key European initiatives such as AI Factories, the cloud–edge continuum, and the Apply AI strategy.

The project pursues three main objectives: (1) optimising AI infrastructure through learning-augmented techniques for efficient scheduling, inference, and resource management across heterogeneous systems; (2) developing secure and trustworthy agentic AI platforms with built-in observability, policy control, and continuous validation mechanisms; and (3) applying these advances to real-world use cases, including industrial robotics, edge AI, energy-aware large language model inference, and AI-assisted scientific workflows.

These objectives will be achieved through an integrated and interdisciplinary approach combining artificial intelligence, cloud-native systems, high-performance computing, and cybersecurity. The work is structured into three core research work packages supported by strong collaboration between academia and industry. A central mechanism is the large-scale exchange of staff across sectors and countries, enabling knowledge transfer, co-development of solutions, and access to cutting-edge infrastructures such as EuroHPC AI Factories and industrial cloud platforms.

The methodology relies on hybrid approaches that combine machine learning with classical optimisation, ensuring both adaptability and robustness in production environments. The project directly contributes to European priorities in trustworthy, sustainable, and deployable AI, strengthening research capacity and accelerating uptake in industry and society.

Topics

Who is involved

OrganisationCityCountryRole
Universitat Rovira I VirgiliTarragonaSpainCoordinator
Universitaet InnsbruckInnsbruckAustriaParticipant
Universita Della Svizzera ItalianaLuganoSwitzerlandParticipant
Universite De NeuchatelNeuchatelSwitzerlandParticipant
Zurcher Hochschule Fur Angewandte WissenschaftenWinterthurSwitzerlandParticipant
Julius-Maximilians-Universitat WurzburgWuerzburgGermanyParticipant
Universitat BayreuthBayreuthGermanyParticipant
Inlecom Innovation Astiki Mi Kerdoskopiki EtaireiaAthinaGreeceParticipant
Barcelona Supercomputing Center Centro Nacional De SupercomputacionBarcelonaSpainAssociatedpartner
Telefonica Investigacion Y Desarrollo SAMadridSpainAssociatedpartner
Universidad De MurciaMurciaSpainParticipant
Universidad De ZaragozaZaragozaSpainParticipant
Universitat Politecnica De CatalunyaBarcelonaSpainParticipant
Institut Mines-TelecomPalaiseauFranceParticipant
Emc Information Systems International Unlimited CompanyOvensIrelandAssociatedpartner
Ibm Ireland LimitedDublinIrelandAssociatedpartner
Stichting VuAmsterdamNetherlandsParticipant
Akademia Gorniczo-Hutnicza Im. Stanislawa Staszica W KrakowieKrakowPolandParticipant
Inesc Id Instituto De Engenhariade Sistemas E Computadores Investigacao E Desenvolvimento Em LisboaLisboaPortugalParticipant
Imperial College of Science Technology and MedicineLondonUnited KingdomParticipant
University of Newcastle Upon TyneNewcastle Upon TyneUnited KingdomParticipant
International Business Machines CorporationArmonkUnited StatesAssociatedpartner

Source: CORDIS, European Commission. Data of 5 Nov 2026, reused under the Commission's reuse policy. View on CORDIS · Grant DOI