Securing IoT and AI systems: A Control Engineering Perspective on Threats and Resilience Strategies from the Cyber to the Physical World

AI has started to revolutionise our lives by allowing enhanced performance, adaptation, and personalisation. At the same time, the IoT is at the forefront of a transformation introducing novel capabilities, such as real-time monitoring, situational awareness, and remote control of devices. And this is happening everywhere around us, from our smart phones and smart devices at home, to the smart meters for a more sustainable grid, till the wearable devices that monitor our health. However, can these technologies be reliably and safely deployed for the control of safety-critical systems? What happens if the information they use is not reliable or accurate? In this session, we will understand the critical importance of identifying vulnerabilities in AI-based and IoT systems, and we will advocate for a systems approach for the design of secure and resilient cyber-physical systems.
 
The Why:
AI and communication technologies are everywhere around us. If you want to take the most of it and ensure that technology does not overpower you, then you need to know more about novel potential safety and security threats and discover novel strategies to make your systems resilient, starting from the design to the operation phases.

Key Takeaways:

• AI and novel communication and sensing technologies offer great opportunities but also open the way to new challenges and vulnerabilities
• A systems perspective is required for safety, security and resilience: it is not possible to separately consider different elements of our systems; everything is nowadays interconnected!
• Control Engineering offers fundamental skills and tools to secure our systems for a safe and resilient future

This session is for... you if you are designing novel products, systems and processes based on AI and IoT technologies, and you want to know more about what to take into account to guarantee safety, security and resilience