#privacy News & Analysis

Federated Learning with Multi-Partner OneFlorida+ Consortium Data for Predicting Major Postoperative Complications

PREFINE: Personalized Story Generation via Simulated User Critics and User-Specific Rubric Generation

Learning to Diagnose Privately: DP-Powered LLMs for Radiology Report Classification

Evontree: Ontology Rule-Guided Self-Evolution of Large Language Models

ChangeNOW Launches Private Send to Break Blockchain Address Tracking

ChangeNOW has launched Private Send, a privacy feature integrated into NOW Wallet that prevents direct address tracking on public blockchains. The feature uses a toggle system to break the link between sender and recipient addresses during transactions.

Aster Chain Launch: Defining a New Era for Onchain Privacy and Transparency

Aster Chain has officially launched, positioning itself as a new blockchain platform focused on combining onchain privacy with transparency. The launch was announced from George Town, British Virgin Islands on March 17th, 2026, marking the platform's entry into the privacy-focused blockchain space.

FedUAF: Uncertainty-Aware Fusion with Reliability-Guided Aggregation for Multimodal Federated Sentiment Analysis

Researchers propose FedUAF, a new multimodal federated learning framework that addresses challenges in sentiment analysis by using uncertainty-aware fusion and reliability-guided aggregation. The system demonstrates superior performance on benchmark datasets CMU-MOSI and CMU-MOSEI, showing improved robustness against missing modalities and unreliable client updates in federated learning environments.

FedPBS: Proximal-Balanced Scaling Federated Learning Model for Robust Personalized Training for Non-IID Data

Researchers propose FedPBS, a new federated learning algorithm that addresses key challenges in distributed AI training including statistical heterogeneity and uneven client participation. The algorithm dynamically adapts batch sizes and applies proximal corrections to improve model convergence while preserving data privacy across distributed clients.

The Landscape of Generative AI in Information Systems: A Synthesis of Secondary Reviews and Research Agendas

Human-Centred LLM Privacy Audits: Findings and Frictions

FedSKD: Aggregation-free Model-heterogeneous Federated Learning via Multi-dimensional Similarity Knowledge Distillation for Medical Image Classification

AudioTrust: Benchmarking the Multifaceted Trustworthiness of Audio Large Language Models

Is privacy still hot? Bitcoin miner Foundry will debut a Zcash pool — here’s why

A writer is suing Grammarly for turning her and other authors into ‘AI editors’ without consent

Anthropic doesn’t trust the Pentagon, and neither should you

AI is capable of remarkable feats. And has the power to kill. Meet one woman warning about the dangers ahead

There Are No Silly Questions: Evaluation of Offline LLM Capabilities from a Turkish Perspective

A study evaluates offline large language models for Turkish heritage language education, testing 14 models from 270M to 32B parameters using a Turkish Anomaly Suite. The research finds that 8B-14B parameter reasoning-oriented models offer the best cost-safety balance for educational use, while model size alone doesn't determine anomaly resistance.

What Makes Code Generation Ethically Sourced?

Privacy by design beats regulation by reaction

The article argues that privacy-preserving decentralized technologies offer a proactive solution to regulatory concerns by addressing centralized digital risks at the design level. This approach reduces systemic vulnerabilities and vendor lock-in compared to reactive regulatory measures targeting centralized systems.

Microsoft Copilot to hijack your browser... for your own convenience

The article title suggests Microsoft Copilot will gain new browser control capabilities, framed as a convenience feature for users. However, no article body content was provided to analyze the specific details or implications of this development.

Cities join Amazon in ending their partnership with license-plate reader Flock following public outcry. ‘Your privacy is totally fine,’ says Ring CEO

Ring CEO Jamie Siminoff comments on public backlash to the company's Super Bowl ad, stating Ring won't use Flock license-plate reader technology. Cities and Amazon are ending partnerships with Flock following privacy concerns from the public.

Can Ethos solve the bot problem? With founder Trevor Thompson

Ethos Network CEO Trevor Thompson discusses how the platform uses social vouching and reputation scores to distinguish legitimate users from bots while preserving user privacy. This approach aims to address the growing problem of bot activity across digital platforms.

The Best AI Tools That Actually Respect Your Privacy

The article reviews nine privacy-focused AI tools as alternatives to Big Tech AI platforms that extensively collect user data. It evaluates different AI tools based on various threat models to help users choose options that better protect their privacy.

Huxe Will Give You a Personalized, Daily Audio Summary Powered by AI

Huxe is a new AI-powered app that creates personalized daily audio summaries by analyzing users' email inboxes and meeting calendars. The app aims to reduce screen time and scrolling, though it raises privacy concerns due to the personal data access required.

TabDLM: Free-Form Tabular Data Generation via Joint Numerical-Language Diffusion

Researchers introduce TabDLM, a new AI framework that generates synthetic tabular data containing both numerical values and free-form text using joint numerical-language diffusion models. The approach addresses limitations of existing diffusion and LLM-based methods by combining masked diffusion for text with continuous diffusion for numbers, enabling better synthetic data generation for privacy and data augmentation applications.