Secure IoT and the Identity of Things

The rapid proliferation of digital technologies and the exponential growth of the Internet of Things have made digital identities the cornerstone of the current digital revolution. As our world becomes increasingly interconnected, the need for secure digital means of authentication has become paramount. However, this presents a significant challenge, particularly for small devices within the Internet of Things ecosystem. This thesis explores three stages in the evolution in digital identities. The first part investigates traditional digital identities, characterized by the use of cryptographic secrets for authentication. Addressing secure storage concerns, researchers have harnessed physical unclonable functions to fortify these identities. The second part explores more sophisticated identity representations, encompassing a group of properties. Verifiable credentials and selective disclosure mechanisms are examined to ensure the privacy and integrity of data. The third part delves into the realm of artificial intelligence as a potential means for flexible identity representation. Large language models emerge as powerful tools for dynamically assembling groups of properties into coherent identities. Combining these AI-enhanced representations into verifiable credentials forms the foundation for a flexible identification framework.