Skip to content Accessibility statement
Institute for Safe Autonomy
HomeInstitute for Safe AutonomyResearchAssurance

Assurance

Assurance is an intrinsic part of safe design, deployment and through-life use of autonomous and AI-enabled systems. 

It’s a principled way of creating ‘justified-confidence’ that a system will operate as expected within a given context through the creation of robust evidence-based safety cases. 

As the capabilities of autonomous systems and AI advance, safety assurance must keep pace with emerging challenges around operation in complex, evolving, and uncertain environments. Assurance must also address ethical and legal concerns and emerging technologies such as generative AI.

The research questions around these issues are the focus of the ISA Assurance pillar. Our work explores a broad range of topics, spans many domains and, critically, is shaped by stakeholders directly responsible for implementing and managing such systems. The result is work which is both practically applicable and scientifically-backed. 

“Safety assurance is the backbone of autonomous and AI-enabled systems. In ISA we have the research capabilities, facilities, and convening power and experience of multi-disciplinary working which is enabling us to deliver expert, impactful support to stakeholders.”

Professor John McDermid, Assurance Pillar Lead

Centre for Assuring Autonomy

ISA’s assurance research is led by the Centre for Assuring Autonomy (CfAA), a research centre within ISA, wholly dedicated to the safety assurance of AI, autonomous systems, and robotics.

SAINTS

The UKRI AI Centre for Doctoral Training in Safe Artificial Intelligence Systems (SAINTS) is a multidisciplinary PhD programme focusing solely on the safety of AI. Research focuses on the lifelong safety assurance of increasingly autonomous AI systems in dynamic and uncertain contexts.