Security and Privacy Aspects of AI-based E-Government Services
The public sector is increasingly adopting AI-based e-government services as they reduce time and costs on the part of government agencies while increasing citizen*s satisfaction. However, such services inherently carry risks, especially when Personally Identifiable Information (PII) is processed. Accordingly, it is important that security and privacy requirements are considered and adhered to both in their development and deployment, otherwise the credibility and trustworthiness of the authorities could be jeopardized.
In traditional software development, a variety of established software development and security testing frameworks already exist for securing services and testing their security. However, since AI is a relatively new and extensive field with various subtopics, it is not clear how to ensure the privacy and security of AI-based applications in practice.
The goal of this study is to analyze what options exist for verifying the security and privacy of AI-based e-government services. In the course of the study, any problems and research gaps will also be identified.