Algorithms that Remember: Model Inversion Attacks and Data Protection Law
This addresses legal and privacy concerns for individuals and regulators by proposing a shift in how models are treated under data protection law, though it is incremental in applying existing security research to legal frameworks.
The paper examines how model inversion and membership inference attacks reveal that machine learning models can reconstruct training data, potentially classifying them as personal data under GDPR, and explores the resulting legal rights and obligations for algorithmic governance.
Many individuals are concerned about the governance of machine learning systems and the prevention of algorithmic harms. The EU's recent General Data Protection Regulation (GDPR) has been seen as a core tool for achieving better governance of this area. While the GDPR does apply to the use of models in some limited situations, most of its provisions relate to the governance of personal data, while models have traditionally been seen as intellectual property. We present recent work from the information security literature around `model inversion' and `membership inference' attacks, which indicate that the process of turning training data into machine learned systems is not one-way, and demonstrate how this could lead some models to be legally classified as personal data. Taking this as a probing experiment, we explore the different rights and obligations this would trigger and their utility, and posit future directions for algorithmic governance and regulation.