Healthcare is seeing rapid growth in AI trials and pilot programmes, but proving that a technology works is only the beginning. Moving from a successful pilot to something that can be safely, effectively and routinely used in clinical practice brings a different set of challenges.  This Data & AI Meetup brings together perspectives from industry, the NHS and academic research to explore what it takes to move AI from experimentation into clinical practice, including the evidence, workflows, trust, regulation and adoption needed to make it work.  Our panel will explore what can help...

Microsoft Virtual Events Powered by Teams

This session forms part of an ongoing series of events in partnership with Scottish Government’s Data Division in which we explore interesting public sector projects and initiatives that highlight and showcase innovative uses of data and AI.     Dr Orian Brook will discuss the findings from her recent analysis of earnings of graduates using the Longitudinal Educational Outcomes dataset for England. Previous work estimated that graduates of creative arts degrees earned less than others, controlling for the institution attended, and personal characteristics of graduates (gender, age,...

AI Security Institute says models by OpenAI and Anthropic went rogue during a cybersecurity test and showed a new type of risk

YouTube video by Hypercube

Agentic AI systems are becoming increasingly autonomous and business-critical, making reliability, safety, and governance essential. Financial services organisations need practical approaches to ensure these systems behave as intended in real-world deployments.  The session will explore practical techniques for financial services assuring Agentic AI, covering evaluation frameworks, guardrails for safe and reliable behavior, and observability practices for monitoring, debugging, and continuous improvement.  Attendees will gain actionable insights and best practices for...