This Trustworthy Artificial Intelligence white paper aims at summarising the main approaches and results obtained by most of the projects that all contributed to the Horizon Europe call HORIZON-CL4-2021-HUMAN-01-01: Verifiable robustness, energy efficiency and transparency for Trustworthy AI.
A Trustworthy AI Cluster has been setup during the project duration to boost the collaboration between the sibling projects. It gathered 9 projects (see the logos on the front page or on top of each chapter page).
Among them, 6 contributed to this white paper:
- EVENFLOW (Robust Learning & Reasoning for Complex Event Forecasting)
- REXASI-PRO (REliable & eXplainable Swarm Intelligence for People with Reduced mObility)
- SAFEXPLAIN (Safe and Explainable Critical Embedded Systems based on AI)
- TALON (Autonomous and Self-organized Artificial Intelligent Orchestrator for a Greener Industry 4.0)
- TUPLES (TrUstworthy Planning and scheduling with Learning and ExplanationS)
- ULTIMATE (mUlti-Level Trustworthiness to IMprove the Adoption of hybrid arTificial intelligencE)
This white paper intends to provide the reader with a testimony of the main achievements of these collaborative projects in terms of AI trustworthiness to go beyond.
The following chapters, organised by project, will provide the reader with a summary of:
- the motivation to address trustworthy AI (whatever the prism is),
- the global approach to introduce the related methodologies and technological solutions,
- the chosen demonstrators with their related results,
- the resulting impacts and perspectives for AI trustworthiness.
[1]M. Barreteau, F. Flammini, N. Katzouris, E. Mezzetti, S. E. Trevlakis, and M. Mattei, “Trustworthy AI white paper”, Sep. 09, 2026, Zenodo.