King's College London

The Trusted Autonomous Systems Hub pulls researchers from computer science and engineering together to develop the trustable autonomous systems of tomorrow. While there is great interest in deploying autonomy in existing and new applications, serious concerns remain about the safety and trustworthiness of current autonomous systems. The Trusted Autonomous Systems (TAS) Hub aims to become a leader in this area, and has identified three key directions:

• Safe and Trusted AI. We consider a system safe when we can provide strong guarantees about the correctness of its behaviour; we say a system is trusted if the average user has confidence in the system and its decision making. The Department of Informatics has a unique portfolio of skill and expertise on model-based reasoning (including verification, norms, planning and argumentation) that can be used to achieve safety and trustworthiness of autonomous systems.
• Explainable AI. There is an urgent need to ensure that the underlying reasoning of autonomous systems is both transparent and comprehensible to those who are expected to act upon them or are otherwise directly affected by them. Providing explanations is key for the practical use of autonomous systems. Researchers in the Department are pioneering new techniques in this domain, such as Explainable Planning and Explainable Security, and there is a joint effort in finding synergies between TAS members to provide better explanations (such as planning and argumentation, or explainable AI and provenance).
• Ethics in Autonomous Systems. In parallel with researching in state-of-the-art model-based techniques for autonomous systems, TAS also focuses on acquiring a deep understanding of ethical, societal, and legal implications of autonomous systems in a research and industrial context, following the Responsible Research and Innovation (RRI) framework.

The Champion of the Trusted Autonomous Systems Hub is Dan Magazzeni, and a full list of members can be found on the Hub website.