Funder: European Commission
Due Dates: March 18, 2027 (Full application, single-stage)
Funding Amounts: €17,500,000 per project (indicative); 2 projects expected to be funded; 100% funding rate for eligible costs
Summary: Funds research on decentralized, federated, and sustainable AI data processing leveraging distributed resources across cloud and edge, with focus on energy efficiency, privacy, and real-world use cases.
Key Information: Participation restricted to entities from EU, Iceland, Norway, Canada, Israel, Republic of Korea, New Zealand, Switzerland, UK, and newly associated countries.
Description
This grant aims to support the development of innovative technologies and methodologies to overcome the limitations of centralized AI computing. The focus is on enabling AI data processing to utilize distributed and heterogeneous resources across the cloud-to-edge continuum during all stages of the AI model lifecycle—including data collection, training, fine-tuning, and deployment. Key research areas include:
- Advanced distributed, decentralized, and federated AI architectures beyond current federated learning.
- Integration of model compression tools and mechanisms to scale AI processing across diverse infrastructures (edge, cloud, HPC).
- Use of novel computing paradigms such as neuromorphic and quantum computing.
- Methods to improve data availability, quality, and privacy in decentralized settings, including cryptographic protections.
- Optimization of data transfers and distributed data management.
- Tools to measure and enhance energy efficiency and sustainability of AI processing.
- Demonstration of these advances in at least two complementary real-world use cases across different domains, with clear value and reproducibility.
The call is part of the Horizon Europe "Digital, Industry and Space" cluster and is designed to address the EU's strategic need for alternative, sustainable AI compute models.
