Algorithms that remember: model inversion attacks and data protection law.

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 indicates 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.This article is part of the theme issue 'Governing artificial intelligence: ethical, legal, and technical opportunities and challenges'.

Direct current stimulation for spine fusion in a nicotine exposure model.

BACKGROUND CONTEXT: Decreased effectiveness in spinal fusion procedures in patients who smoke before, during, or after the operation has been noted in several clinical studies. In previous work, direct current (DC) electrical stimulation has been shown to enhance inter-transverse process fusion in a rabbit model. PURPOSE: To test the efficacy of DC stimulation on bone healing in spinal fusion in rabbits exposed to nicotine. STUDY DESIGN/SETTING: A randomized and controlled interventional study. METHODS: Thirty male New Zealand white rabbits received a single level posterolateral, inter-transverse process fusion with autologous iliac crest bone. One group (control) acted as a control without nicotine or electrical stimulation. A second group (Nic) received a continuous dose of nicotine via a transdermal patch to simulate a heavy smoker, and a third group, nicotine/stimulator group (Nic/Stim), additionally received a 100-microamp DC stimulator. The fusion masses (L5-L6) and the adjacent unfused control segment (L4-L5) were evaluated radiographically, manually, and biomechanically. RESULTS: The Nic group showed significantly higher fusion rate compared with the control group. The Nic/Stim group also demonstrated significantly higher fusion rate and X-ray trabeculation compared with the control group. However, the Nic/Stim group was not significantly higher than the Nic group in fusion rate or X-ray trabeculation. CONCLUSIONS: Nicotine significantly improved fusion rate compared with controls, and DC stimulation significantly increased X-ray trabeculation of nicotine treated rabbits compared with controls.