PhD students from the Centre for Doctoral Training in Safe and Trusted AI – the STAI CDT – C Henrik R Åslund (2019 cohort) and Francis Rhys Ward (2020 cohort) were co-organisers of the 2023 Workshop on Safe and Trustworthy AI (STAI 23), along with Dr Elizabeth Black and Dr Francesco Belardinelli (Director and Co-Director of the STAI CDT).
The STAI 2023 workshop was held on 9 July 2023 at the International Conference on Logic Programming (ICLP) in London and was focused on the broad area of safe and trustworthy AI.
The workshop consisted of accepted papers (presented either in talk or poster form), an invited talk on the adversarial susceptibility of large neural networks, and a fireside chat on the relationship between the AI ethics community and the catastrophic AI risk community. The invited speakers were El Mahdi El Mhamdi, Atoosa Kasirzadeh and David Krueger.
STAI CDT PhD student Richard Willis won the second prize for the best presentation, voted on by attendees, for his paper ‘Resolving social dilemmas through rewards transfer commitments’, co-authored with Professor Michael Luck. The paper can be found here.