Abstract
One might wonder what the Austrian physicist’s famous thought experiment and data classification have in common. In fact, the paradoxical dual state of simultaneous aliveness and death of Schrödinger’s cat, could be used as a practical foundation to exemplify a similar behaviour when distinguishing between personal and non-personal data in the context of synthetic information. The European data protection framework has at its core the binary distinction between personal and non-personal data. However, the current increase in the implementation of synthetic data, meaning algorithmically generated information, poses a challenge to this rigid classification. This is mainly because while synthetic data is often seen as a privacy-enhancing technology, not all synthetic information is the same and the risk of reidentification makes its classification legally ambiguous. Hence, as it will be further illustrated, and much like Schrödinger’s hypothetical cat sealed inside a box – which results both death and alive until the box is opened, data’s categorisation may require a more fluid approach. This opinion paper examines whether the binary model under the GDPR is sufficient to mitigate the adverse impacts of synthetic data. The first chapter exemplifies the nature of this kind of data, as well as the methodologies utilised to generate it, and possible legal challenges related to its use. Next, the limitations of the current EU data protection law framework are highlighted, particularly focusing on its applicability on synthetic data and dynamic data flows. Finally, the paper introduces alternative perspectives to the risk-based approach and binary divide between personal and non-personal data, drawing also from quantum mechanics notions. By critiquing the inadequacy of the current framework considering synthetic data through a critical, and interdisciplinary technology-focused legal lens, this paper argues data protection law must evolve beyond static classification, focusing instead on the ever-evolving status of data. At the same time, it recognises further discussion is still needed within a field that is relatively novel.
