GCA

Work Description:
I’m excited about contributing to the re-decentralisation of the Internet. I am against giving away the control of the Internet to the big online monopolies, who have been increasingly and unnecessarily centralising it over the last 10 years. I am developing decentralised technologies to prove that there are alternatives. A city is decentralised when, to some extent, it is free from monopolies that gather, own and control large volumes of data that the city produces. I feel strongly against and deviate from the traditional centralised model that sooner or later leads to unduly data monopolisation. My research focuses on the development of Privacy Enhancing Technologies (PETs) for decentralised application integrations. These integrations can collect and process data produced by independent companies to provide value while adhering to strict data protection requirements such as privacy, non-exfiltration, non-dissemination and non-monopolisations. PETs are an abstract concept that covers a broad range of complementary and competing technologies which can be used to solve different problems, such as, for building Secure Private Computing-as-a-service. Trusted Execution Environments (TEEs) are one of the most promising PETs and are the main focus of my research. My objective is to design algorithms and build software tools to demonstrate that privacy-aware decentralised application integrations can be implemented using TEEs in both software or hardware. Software-based TEEs utilise recent advances in cryptography. Emerging techniques in this area include: Differential Privacy (DF), Secure Multi-Party Computation (SMPC), and Homomorphic Encryption (HE). My task is to evaluate these techniques to determine whether they are ready for building practical applications. Hardware-based TEEs are more mature, with several competing technologies available in the market. The TEEs under evaluation include: Intel SGX, ARM trustzone, AMD virtualisation, Amazon AWS Nitro Enclaves AVAILABLE FROM THE CLOUD, and attestables built on the basis of hardware capabilities included on the Morello Board platform released by ARM for evaluation purposes in 2022.