Trust By Decentralization? Where Blockchain Systems Meet Responsible AI Challenges
Aggelos Kiayias¹
DOI : 10.65701/v1c7t4p0z2
Corresponding authors:
aggelos.kiayias@ed.ac.uk
Abstract
The importance of developing “Responsible AI” has recently soared as one of the key considerations in information technology (IT). It is thus pertinent to ask what insights one may glean from other research areas within computer science that have been focusing on solving related problems or building tools and methods for increasing the trustworthiness of IT systems and can be repurposed within this new context. Blockchain technology can be seen to intersect with the objectives of responsible AI and can be used as a foundational infrastructure for responsible AI development. To highlight this direction, we briefly outline four use cases: decentralized identity for AI systems, blockchain-based governance, privacy-enhanced data mining, and decentralized machine learning. We conclude by discussing a “rights by design” approach for responsible AI development and with an overview of the papers in the session.
