Privacy Intrusion and National Security in the Age of AI

This CETaS Research Report explores the complex issue of privacy intrusion arising from the use of automated analytics, with specific focus on artificial intelligence (AI). The research focuses on UK national security and law enforcement agencies with access to legal powers that incur some degree of intrusion into individuals’ private lives. As automated methods are increasingly deployed to process the data collected through the use of such powers, there is a need to understand the additional privacy considerations that could arise as a result of this automated processing.

The report’s ultimate objective is to develop a structured analytical framework for assessing proportionality of privacy intrusion arising from the use of automated analytics. The framework considers the whole lifecycle of automated analytics, including data collection, training, testing processes, and use. It aims to introduce a common language and taxonomy that will assist stakeholders in identifying, comparing, and assessing the potential impact of relevant privacy considerations in a structured and evidence-based way. This framework is not intended to replace any existing authorisation or compliance processes, but rather to provide an additional layer of rigour and assurance to supplement and futureproof existing processes.

The research is informed by semi-structured interviews and focus groups with stakeholders across the UK government, national security and law enforcement, and legal experts outside government, as well as an understanding of the literature on proportionality in English law and critiques of the application of the proportionality test. Particular attention is paid to the distinctive aspects of automated processes and artificial intelligence, and the requirements for making a structured analytical framework useful in practice.