Personalization Meets Safety:Mechanisms,Risks,and Mitigations in Personalized LLMs
2026-06-08 • Artificial Intelligence
Artificial Intelligence
AI summaryⓘ
The authors review how Large Language Models (LLMs) are personalized to users and the new safety risks this creates. They organize personalization methods and user representations, then identify specific vulnerabilities and ways to reduce risks. They find current research has gaps, like ignoring how safety changes with different users and not considering combined personalization methods. Their work provides a unified view to help build safer personalized LLMs and suggests future research directions.
Large Language ModelsPersonalizationUser RepresentationSafety RisksPromptingFine-tuningReinforcement LearningMixture-of-ExpertsEvaluation Methodologies
Authors
Yanyan Luo, Xue Han, Ruiqiao Bai, Xin Huang, Yitong Wang, Qian Hu, Qing Wang, Chunxu Zhao, Jie Liu, Cong Geng, Lehao Xing, Pengwei Hu, Junlan Feng
Abstract
Large Language Models (LLMs) have enabled increasingly personalized interactions by adapting to users' preferences, contexts, and long-term histories. However, the mechanisms that enable personalization also expand the safety landscape in ways not systematically addressed by existing literature. Existing reviews typically focus either on personalization or safety, leaving their intersection largely unexplored. We present the first comprehensive, safety-aware review of personalized LLMs. We organize personalization along three dimensions-user representation, personalization paradigm, and evaluation-and introduce a unified taxonomy of safety risks. At the representation level, we analyze risks arising from diverse user representations. Across mainstream personalization paradigms, we delineate vulnerabilities inherent to prompting, retrieval augmentation, parameter fine-tuning, reinforcement learning, Mixture-of-Experts (MoE), pruning, agent frameworks, and multimodal personalization, and synthesize mitigation strategies across the model lifecycle. Beyond these fine-grained risks, we characterize paradigm-agnostic safety risks arising from personalized adaptation. We further summarize personalized datasets and evaluation methodologies. Through a case study of OpenClaw, we analyze deployment trends in personalized agent ecosystems. Our analysis reveals three structural inadequacies in existing research: safety is evaluated as user-invariant rather than relational, personalization techniques are analyzed in isolation rather than in composition, and evaluation frameworks cannot capture emergent long-term risks. By jointly examining personalized representations, personalization paradigms, safety risks, defenses, and evaluation methods, we provide a unified framework for developing safe personalized LLMs and highlight key directions for future research.