Files
Abstract
In recent years, law enforcement agencies have used artificial intelligence to predict crime both domestically and internationally. With the development of technology also comes changes to the legal landscape to protect an individual’s privacy. The need to regulate privacy is indisputable, so a mass surveillance system built with standards to protect privacy can help law enforcement incarcerate or exonerate individuals. This thesis argues that a well-designed surveillance system can achieve this balance using state-of-the-art convolutional neural networks and a human-in-the-loop architecture. Once technical and ethical concerns have been addressed and appropriately resolved, mass surveillance systems can help policing agents exonerate the innocent, hold individuals responsible for their actions, and keep societies safe.