This WP aims at establishing a continuous interdisciplinary dialogue for investigating the methods and methodologies to design, develop, assess, enhance systems that fully implement Trustworthy AI with the ultimate goal to create AI systems that incorporate trustworthiness by design. This activity will be organized along the six dimensions of Trustworthy AI: explainability, safety and robustness, fairness, accountability, privacy, and sustainability. Each task aims at advancing knowledge on a specific dimension and puts it in relationships with foundation themes. The overall mission for Trustworthy AI is to combine the various dimensions in the TAILOR research and innovation roadmap. Moreover, to maximize this overall goal and take advantage of any effort in Europe, TAILOR will also interact and collaborate with the activities related to “AI Ethics and Responsible AI” of the proposal Humane-AI-net and will lead the organization of joint scientific actions. The overall activity is articulated in three main steps:
setting the stage for significant research challenges, putting in connection the various dimensions of Trustworthy AI with foundational themes for the definitions of requirements, demands, focused gaps, and understanding inter-dependencies;
leveraging the development of foundations and technologies for the various dimensions of Trustworthy AI with joint focused research;
moving from Trustworthy AI guidelines towards their instantiation into concrete standardized processes for developing and maintaining AI systems that incorporate the results of the foundational research to properly address trust by design and fuel the Strategic Research and Innovation Agenda on Trustworthy AI with input gained by the above activities.
TAILOR is an ICT-48 Network of AI Research Excellence Centers funded by EU Horizon 2020 research and innovation programme GA No 952215.