IDL designs the User Experience Design handbook of the AIXPERT Horizon Europe project.
The AIXPERT project (AI eXplainability, PERformance, and Trustworthiness through Agentic Frameworks) is a Horizon Europe Research and Innovation Action (Grant Agreement No. 101214389, HORIZON-CL4-2024-HUMAN-03). It focuses on developing explainable, accountable, and transparent AI systems that can be trusted, audited, and human-centred across different application domains.
AIXPERT introduces an adaptive, situation-aware AI-agentic platform that integrates multi-agent systems, multimodal foundation models, and real-time human feedback (RLHF). Its core ambition is to improve AI explainability, transparency, robustness, and autonomy while maintaining flexibility across architectures and use cases. The system features three main layers:
Agent-World Interface Layer – coordinates agents and their real-world awareness.
Dialogue Mediation Layer – manages agent–user and agent–agent communications.
Cognitive Foundation Layer – provides explainable, multimodal AI capabilities.
AIXPERT will demonstrate its innovations in five pilot sectors: healthcare, recruitment, manufacturing, educational robotics, and the creative industries (e.g. music). Each pilot evaluates trust, transparency, usability, and societal impact of explainable AI applications.
The project’s main objectives include:
Enhancing AI explainability and accountability across complex workflows.
Developing a framework for trustworthy AI evaluation.
Supporting EU AI policy and standardisation through actionable recommendations.
Advancing human-centric AI design to foster ethical, fair, and inclusive AI systems.
Expected outcomes:
A modular AIXPERT platform with integrated AI Assistant and agentic tools.
A No-Code framework for configuring and monitoring AI workflows.
Cross-sector pilots validating societal and economic benefits.
Policy and standardisation recommendations aligned with EU AI Act priorities.
