Browsing by Author "Gunst, Helena"

The decisions of everyday life are to an increasing extent made by a new deciding force: the proprietary algorithm. The information society of today is a “world of automatic decision-making”, as more and more decisions are delegated to automatic systems which are able to process data and make decisions, often with little supervision from human decision-makers. Such automated decision-making presents an interesting dichotomy of interests: the companies that develop the algorithms and data processing mechanisms have an interest in keeping them protected and hidden from competitors, while individuals have a vested interest in getting insight into, and understanding of, how decisions based on their personal data are being made. This thesis sets out to analyze this conflict, looking to answer how data protection rights to a transparent processing of personal data can be reconciled with trade secret rights in automated decision-making. The topic is studied from an EU point of view with legal dogmatism as the methodological base, and the main research question is divided into three parts corresponding with the chapters of the thesis. Firstly, the data protection rights to a transparent processing of personal data are as regards automated decision-making are reviewed. The main rights relating to the transparent processing of personal data are found in the notification obligations, access rights and safeguards relating to automated decision-making of the upcoming General Data Protection Regulation, and the rights are also collectively referred to as the “right to explanation”. The analysis in the chapter shows multiple ambiguities and gaps, leaving the scope of the rights unclear and to be further clarified in case law and through guidance by data protection authorities. The multiple requirements and the ambiguous nature of the right to an explanation results in that it allows both data subjects and trade secret holders to read the right in a favorable way to their own interests, giving rise to a significant legal uncertainty. Secondly, the trade secret rights behind automated decision-making are analyzed. Trade secret law offers a wide protection for various types of data processing algorithms and methods a private company may use in the automated processing of personal data. As regards the protection of such information in the light of data subject transparency rights, the most crucial interest of trade secret holders is to ensure that sufficient steps are taken to keep the information secret. This interest is jeopardized by the obligations on companies to meet the requirements of data subject transparency rights, since a disclosure to a large amount of data subjects could render the information generally known, losing its trade secret status. While trade secret rights may function as one of companies’ main ways of limiting the applicability of the right to explanation, the General Data Protection Regulation also makes clear that trade secret rights cannot be the base for refusing data subjects all information. Thus, data controllers must walk a fine line between meeting their data protection obligations while simultaneously keeping their data processing methods secret. Lastly, a balancing exercise is conducted in order to analyze how the conflicting rights in automated decision-making can be balanced and reconciled. The analysis of the balancing norms of the Trade Secrets Directive and the General Data Protection Regulation shows a contradictory situation that affirms both a precedence of trade secrets on data protection rights, as well as a precedence of data protection rights on trade secrets. Ultimately, a slight preference for data protection rights is seen. However, national legislation and case law on the previous right of access under the Data Protection Directive shows another reality where trade secret rights have a strong restricting factor on any transparency rights of data subjects. It remains to be seen whether the reinforced rights of the General Data Protection Regulation might embolden national legislators and courts to implement a more transparency-friendly approach in which trade secrets are given less protection when compared to the right to an explanation. Until then, the possibilities of the right to an explanation to enhance transparency in automated decision-making must be viewed with certain disbelief, due to the ambiguity of the right itself and the strong restricting nature of trade secret rights. As a consequence, solutions for reconciliation must be found elsewhere, and the thesis concludes by presenting three alternative approaches to reconciling the rights in order to enable a functioning transparency in automated decision-making.