Data Privacy

Apple's macOS appears to include a hidden classification feature that automatically identifies family relationships between people in photos and personal data, as part of Apple Intelligence's personal knowledge graph. Users and developers discussing it note the capability was never documented by Apple, raising questions about how the company profiles personal relationships on-device and what else may be quietly built into every Mac.

Italy's competition authority has fined Meta 887 million dollars for allegedly forcing Android users to link their accounts and granting the company access to their data. The penalty, reported by Reuters, is one of the largest privacy-related fines issued against a major technology firm in Europe and targets Meta's data collection practices on its Android platform.

The Electronic Frontier Foundation is urging an appeals court to rule that police must obtain a warrant before searching automated license plate reader (ALPR) databases. The group argues that these systems track drivers' movements on a massive scale, and that unrestricted searches of the collected data violate constitutional privacy protections.

A new essay by developer Rakhim Diafutdinov makes an unusual argument: instead of the usual complaints about surveillance, the author wishes tech companies knew more about him. The piece argues that better knowledge of users could mean fewer irrelevant ads, less junk content and services that actually fit people's lives, a contrarian take in a debate usually framed entirely around privacy.

A question on Hacker News asks users to recommend television sets with privacy in mind, drawing hundreds of responses. The discussion centers on smart TVs' data collection, including built-in tracking, automatic content recognition, and microphones, with users suggesting ways to disable telemetry or opting for dumb monitors instead.

The New York Times reports that Meta has been ordered to pay $25 billion for deceiving users over privacy. The ruling marks one of the largest penalties tied to data privacy practices, and the case is drawing wide attention to how the company handled user data and what it told people about it.

Anthropic has launched Personal Finance for Claude, allowing the AI assistant to connect to bank accounts and cards and analyse users' spending. Italian tech outlet Punto Informatico reports the news, detailing what data Claude can access and what the feature cannot do, drawing attention to privacy and financial data questions.

Revolut is reportedly shifting to a self-hosted deployment of GLM, the large language model developed by Z.ai (formerly Zhipu AI). The fintech company, one of Europe's largest banks, appears to be bringing AI inference in-house rather than relying on external providers. The move is drawing attention among developers discussing cost, control and data-privacy trade-offs of self-hosting frontier Chinese open models.

A widely shared tech story is drawing attention to the most privacy-focused version of Android, recounting the little-known history behind how it was built and why it stands apart from mainstream Android. Interest centers on growing public concern over smartphone privacy and data collection, with viewers engaging heavily with the account of the operating system's development and its promise of stronger user protections.

French electronics retailer Fnac Darty has been fined €7 million following a complaint pursued by privacy group noyb, founded by activist Max Schrems. The penalty stems from violations of EU data protection rules. It marks another enforcement success for noyb, which has built a record of complaints against major companies over how they handle user data under GDPR.

Ford has introduced the Ford Privacy Promise, a commitment stating that customers remain in control of their personal data. The announcement appears on Ford's own publication, From the Road, framing data privacy as a customer right rather than a corporate courtesy. Early reaction is measured, with readers scrutinising how much practical control drivers actually gain and what data Ford's connected vehicles continue to collect.

Personal finance experts are warning people about sharing sensitive financial information with ChatGPT, saying at least one category of detail should never be disclosed to the chatbot. The advice reflects growing concerns about privacy and data security as more people use AI assistants for budgeting, investing, and money questions. No specific detail was named in the headline itself.

A Swedish authority has ruled that Lexbase, a service that lets users search for personal data about individuals, is not permitted to make people searchable. The decision, reported by SVT Nyheter, is drawing attention in Sweden because it limits the database service's core function and touches on ongoing debates about privacy and public access to personal information online.

A cybersecurity explainer is warning that covering or blacking out a name in a PDF does not necessarily delete the underlying data: in many cases the original text remains recoverable, exposing sensitive information. The discussion outlines why common redaction methods fail and which proper practices should be followed instead, urging people to use tools that genuinely remove content rather than simply obscure it visually.

DuckDuckGo is marketing its privacy-focused search service to schools, aiming to capitalize on growing concern over Google's AI policies in education. The move comes as backlash builds against Google's AI features being embedded in tools used by students, with schools increasingly looking for alternatives that limit AI exposure and data collection.

Brazil's national telecommunications agency is moving to make facial biometrics a requirement for mobile phone users. According to GAZETA, the measure would oblige customers to register facial identification when acquiring or maintaining mobile lines. The proposal is drawing attention amid wider debates about data protection, government surveillance and identity verification rules in the country's telecom sector.

A prospective TV buyer who has not owned a set in years is asking which large-screen models respect privacy, seeking a 60- to 80-inch 4K display with at least 60Hz. The question highlights growing concern over smart TVs tracking viewing habits and serving ads, as people look for sets that can be used without constant data collection.

The Baden-Württemberg Commissioner for Data Protection and Freedom of Information (LfDI) is launching its AI Week with a 3 pm keynote by internet policy researcher Ralf Bendrath. His talk, 'Digital rights thrown under the bus', argues that the EU's reform agenda threatens to damage data protection. The week-long programme asks which questions artificial intelligence can actually answer.