by Iraklis Symeonidis, COSIC, KU Leuven
and Gergely Biczók, CrySyS Lab, BME
Recent technological advancements have enabled the collection of large amounts of personal data at an ever-increasing rate. Data analytics companies such as Cambridge Analytica (CA) and Palantir can collect rich information about individuals’ everyday lives and habits from big data-silos, enabling profiling and micro-targeting of individuals such as in the case of political elections or predictive policing. As it has been reported at several major news outlets already in 2015, approximately 50 million Facebook (FB) profiles have been harvested by Aleksandr Kogan’s app, “thisisyourdigitallife”, through his company Global Science Research (GSR) in collaboration with CA. This data has been used to draw a detailed psychological profile for every person affected, which in turn enabled CA to target them with personalized political ads potentially affecting the outcome of the 2016 US presidential elections. Whether CA used similar techniques at the time of the Brexit vote, elections in Kenya and an undisclosed Eastern European country (and several other countries) is under investigation. Both Kogan and CA deny allegations and say they have complied regulations and acted in good faith.
This blog post does not take sides in this debate, rather, it provides technical insight into the data collection mechanism, namely collateral information collection, that has enabled harvesting the FB profiles of the friends of app users. In this context, the term collateral damage refers to the privacy loss these friends suffer. In a larger nexus, this issue is part of interdependent privacy when your privacy depends unavoidably on the actions of others (in this case, your friends). Continue reading Interdependent privacy in effect: the collateral damage of third-party apps on Facebook