Original publication (PDF, Danish, Google Drive link from the NGO): https://drive.usercontent.google.com/download?id=1MZrFRii_nJYdW8RulORB9JveLkCRbncX&export=d

Original publication (PDF, Danish, Google Drive link from the NGO): https://drive.usercontent.google.com/download?id=1MZrFRii_nJYdW8RulORB9JveLkCRbncX&export=download&authuser=0 https://drive.usercontent.google.com/download?id=1MZrFRii_nJ... via their Google Drive link at: https://www.digitaltansvar.dk/viden/analyser https://www.digitaltansvar.dk/viden/analyser The article left me wondering whether they actually cultivated a network of kids interested in self-harm. It sounds from the study (such as it is) that they in fact just simulated vulnerable 13-year-olds’ accounts. Which comes as something of a relief from an ethical point of view. I wonder, though, if that method may have confounded their findings. I could imagine Instagram’s recommenders linking their simulated 13-year-olds’ accounts based on signals indicating their common association with the research organization (rather than purely on the basis of their shared interest in self-harm). Maybe somebody with stronger Danish than mine could shed light?