publications – Preliminary Work
Determining how voice assistants should broker consent to share data with third party software has proven to be a complex problem. Alexa now offers “voice-forward consent”, allowing users to grant skills access to personal data mid-conversation using speech. While more usable and convenient than current alternatives, asking for consent 'on the fly' can undermine several concepts core to the informed consent process. The intangible nature of voice interfaces further blurs the boundary between parts of an interaction controlled by third-party developers from the underlying platforms. We outline a research agenda towards usable and effective voice-based consent to address the problems with brokering consent verbally, including our own work drawing on the GDPR and work on consent in Ubicomp. Read More ›
recruitment – Position Now Closed
This two-year post will focus on symbolic and norm-based AI models to specify, dynamically monitor, and explain the desirable behaviour of AI assistants. Read More ›
recruitment – Position now closed
We are seeking to hire an outstanding Research Assistant/Associate to join the Verification of Autonomous Systems (VAS) group, led by Prof. Alessio Lomuscio. The VAS group is a friendly, vibrant, multi-national team working on various aspects of safe Artificial Intelligence, including verification and safety aspects of machine learning systems, autonomous systems, swarm-based systems, robustness and safe reinforcement learning. The group also contributes to US-funded projects on Assured Autonomy and has strong links with the UKRI Centre for Doctoral Training in Safe and Trusted Artificial Intelligence. Read More ›
funding – Grants & Funding
Led by King’s, a collaboration between King’s Cybersecurity Centre, Imperial College London and non-academic partners has received £1.5 million from the Engineering and Physical Sciences Research Council (EPSRC) to carry out pioneering research to further understanding of attacks on Secure AI Assistants. Read More ›