Making AI Socially Safe

Event details

In this talk, TT Arvind will be discussing the challenges of creating AI systems that are socially safe and socially trustworthy. He will draw on emerging real-world examples of the use of AI in contexts as diverse as policing, medical triage, social care, and other forms of decision-making. For the use of AI to be trustworthy in any of these contexts, we require systems that can be trusted to produce outcomes that do not conflict with accepted principles of ethics, human rights, and wider socio-cultural expectations and norms. Arvind will show why top-down, external regulation is unlikely to provide a satisfactory answer to these challenges. Drawing on the work of the TAS Resilience Node, he will show how the solution requires these norms to be adapted to and embedded in every stage of the AI lifecycle, including specification, development, evaluation, and assurance; and will discuss current developments and the state of the art in meeting this challenge.

After the talk, we invite you to join us for a networking session from 2pm-3pm (in-person only). This is an excellent opportunity to engage with our speakers, fellow attendees, and industry professionals.

This event is led by York Law School and supported by the School for Business and Society.

Please note that this is a hybrid event. If you have booked an online ticket, you will be sent the Zoom link the morning of the event.

About the speaker

TT Arvind

TT Arvind is Professor of Law and Head of Department at York Law School. He is a co-investigator on the UKRI Trusted Autonomous Systems Resilience Node, which brings together academics and industry to devise new ways of creating AI systems which are trustworthy and resilient.