Safe and Efficient Frontier AI
Funds a €44M frontier-AI project that builds safer, more efficient European models with lower computational demands.
Frontier AI models could unlock major scientific and economic opportunities, but their energy use, computational cost and safety risks are growing quickly. This topic funds European research that advances frontier-model capability while tackling those constraints directly.
The project can develop and train a state-of-the-art multimodal model with tool use, reasoning and autonomous problem-solving…
Advance a sovereign European frontier AI model toward state-of-the-art performance with multimodal and agentic capabilities, while developing methods that reduce computational demand and make advanced AI safer. Research may cover compression, distillation, efficient architectures, lower-cost training and inference, automated safety testing, interpretability, misalignment, reward hacking, encoded reasoning, adversarial attacks, jailbreaks, backdoors and agentic-AI risks. Start at TRL 2 and reach TRL 4, involve the European AI research community, attract talent, connect with RAISE, AI Factories and Data Labs, and allocate cohesion work with ADRA and the GenAI4EU central Hub.
Sign up free to see the funding breakdown
Sign up free to see the industries in scope
Sign up free to see the full eligibility
Sign up free to see how to apply
Sign up free to see what you submit
Sign up free to see how they score you
Sign up free to see the timeline
Sign up free to see where teams trip up