Workshop on Trustworthy Artificial Intelligence

There are several dimensions that concur to create a trustworthy AI system, like the capability of being explainable, safe and robust, able to guarantee fairness, equity and justice, accountable and reproducible, respectful for privacy, and sustainable. 

Fundamental research is required to combine all these dimensions in a principled way and for coping with properties and tensions among conflicting dimensions. Controlled experimentation environments are also crucial to enable researchers to play with their system, assess their level of trustworthiness and repair them if needed. 

By developing synergies and cross-fertilization between practitioners and researchers, this workshop will conduct a reflection on the methods and principles allowing the integration of artificial intelligence in critical products and services in a safe, reliable and secure way. Particular attention will be paid to methods for the engineering of innovative industrial products and services integrating AI, the large-scale deployment of industrial systems, as integrating AI is a crucial issue for industrial competitiveness. 

The scientific challenges include, as example, the construction of AI components with controlled confidence, the construction of data/knowledge to increase trustworthiness in learning or the interaction generating confidence between the user and the AI-based system. 

Authors are invited to submit their original work in this field and thus contribute to the reflections that will be conducted. 

The workshop will include oral and poster presentations and will leave a lot of room for exchange and discussion between participants.