On 28 September 2026, at Royal Society Publishing hosted ‘Safe, secure and robust AI for safety-critical systems‘, a virtual seminar in its ‘Behind the paper with Royal Society journals‘ series.
Authors presented work from the theme issue ‘Safe, secure and robust AI for safety-critical systems‘ in Philosophical Transactions of the Royal Society A, compiled and edited by Ajitha Rajan and Desmond J. Higham. The talks asked whether AI can safely be deployed in safety-critical systems and whether AI systems can be secure, and examine prompt engineering as a way to mitigate medical bias in large language models.
SPEAKERS
Tamara G. Kolda – MathSci.ai: ‘Is it safe to deploy AI in safety-critical systems?’
Desmond Higham – University of Edinburgh: ‘Can AI systems be secure?’
Ying Xiao – King’s College London: ‘Mitigating medical bias in large language models by prompt engineering: an empirical study of effectiveness and trade-offs’
CHAIR
Ajitha Rajan – University of Edinburgh
#AISafety #AI #LLMs #PhilTransA