Human Rights, Regulation and the Development of Algorithmic Police Intelligence Analysis Tools in the UK

Abstract

The deployment of machine-learning technologies around decision-making and decision-support contexts in UK policing raises some new, and in other ways some very familiar, human rights issues. The discussion in this paper is centred around key considerations of the right to restriction of law enforcement processing (of data points in an algorithm) where the data processing may lead to 'inaccuracy' that is misleading - and which in turn might lead to errors in the treatment of a data subject in the justice system that may be a breach, not just of data protection rights, but of fundamental human rights, as a result. This paper also gives a short case study, from a human rights perspective, on the trials of facial recognition technology by a police force in Wales (UK).

Keywords: human rights, regulation, algorithms, machine learning, trade offs, specificity, sensitivity, data protection, big data, surveillance, police, law, justice