6 articles in this topic.
Privacy risks in modern AI systems are not limited to obvious data leaks. They also appear through indirect channels such as membership inference, memorization extraction, and retrieval traces that expose sensitive context.
These articles analyze how privacy leakage happens in practice and what engineering controls reduce exposure. You will find discussions on threat modeling, evaluation methods, and lightweight safeguards that can be integrated into existing model and RAG deployments.
This page is maintained as a high-signal index for Privacy. Use it to follow newer articles first, then branch into adjacent topics and defensive patterns that repeatedly appear across projects and paper reviews.
The dominant security theme today is the structural breakdown of boundaries between reasoning engines and executive environments, transitioning the primary threat vector from semantic prompt manipulat
The systematic scaling of automated, AI-driven vulnerability discovery has triggered a structural crisis in legacy patch-management frameworks, as evidenced by the 263% surge in CVEs forcing an overha
A novel zero-shot machine unlearning method using information theory and curvature analysis, enabling efficient removal of data influence without requiring access to the retain set.
An introduction to machine unlearning in Large Language Models, covering the TOFU benchmark, various unlearning methods (GradDiff, NPO, IdkPO, AltPO), and the challenges of maintaining model utility while forgetting specific knowledge.
A comprehensive analysis of membership inference attacks against RAG systems, examining three state-of-the-art approaches: RAG-MIA, S²MIA, and MBA, along with their defenses and limitations.
An analysis of neural phishing attacks that teach LLMs to memorize and leak private information by inserting benign-appearing poison data during pretraining, achieving up to 90% secret extraction rates.