Safety v Security

Challenges and applications in the cyber security era

On behalf of the Cyber Security Centre at the University of Southampton I’ve organised, supported by the UK National Cyber Security Centre, the first workshop on Safety v Security.

The workshop gathered more than 70 experts across academia, government and industry providing insights on pressing security challenges in modern interconnected cyber-physical safety-critical systems.

The workshop touched on a large variety of applications, including automotive, avionics and CNI, and testified the key role safety and security can have in the development of reliable, trustworthy modern systems.

Some of the slides and interviews to the speakers can be found on the cyber security group web page. Below a chat with two of the distinguished speakers.

Throughout the talks and the networking sessions, it emerged that we must stop addressing safety and security as different — alternatively, stop giving safety priority over security, or vice versa. Both have a common target: ensuring systems will work and continue working as expected.

Given the advent of new technology, the changing landscape of cyber threat actors and cyber attack methods: we must prepare to address threats as they emerge. All threats cannot be predicted at design time and the changing operating environment of modern computing systems impose to systems to be able to adapt over time against emerging threats.

Last but not least, as AI appears to be a game changer and to be more and more embedded into cyber-physical systems, the key concern for security and safety of next generation autonomy systems is to develop principled design, development and monitoring approaches for AI applications.