AI-DAPT Participates in Data Week 2026 Session on Trustworthy and Efficient AI Pipelines
Last week, the AI-DAPT project participated in Data Week 2026, joining the session “When Data Pipelines Decide AI Trust and Efficiency: theory and practice”, which brought together several European initiatives working on trustworthy, data-centric, and efficient AI systems.
The session gathered projects funded under the European Commission topic “Efficient trustworthy AI – making the best of data”, including PANDORA Horizon Europe Project, AI-DAPT, MANOLO Project, EXTRA-BRAIN, RAIDO Project, as well as related initiatives such as the Turing Project. The discussion provided an opportunity to exchange perspectives on how AI pipelines can move beyond isolated experimentation and support reliable, scalable, and transparent AI deployment in real operational environments.
Representing AI-DAPT, Carlos Agostinho (UNINOVA) presented the project’s vision for trustworthy AI, emphasising that trustworthiness should not be viewed solely as a high-level principle or regulatory objective, but as something that must be embedded directly into the design and operation of AI systems and pipelines.
The presentation highlighted how AI-DAPT approaches trustworthy AI through the continuous orchestration and monitoring of data and AI pipelines, integrating observability, validation, adaptation, and human oversight across the AI lifecycle. In this perspective, trustworthy AI becomes an operational characteristic of the pipeline itself — supported by transparent data handling, continuous validation, explainability, and adaptive mechanisms capable of responding to evolving conditions and requirements.
The session also explored broader challenges related to AI efficiency, reproducibility, and data governance, reinforcing the growing importance of robust AI infrastructures capable of supporting trustworthy and scalable AI across domains.
AI-DAPT was pleased to contribute to this discussion and exchange ideas with fellow researchers and experts, including Nuria De Lama, Maria Pateraki, Konstantinos Tserpes, Julia Palma, and Christos Vasilakis.
We would like to thank the organisers, speakers, and participants for the engaging session and valuable discussions. Events such as Data Week continue to play an important role in strengthening collaboration across the European AI ecosystem and advancing the development of adaptive, transparent, and trustworthy AI systems.
