GrandGuard: Taxonomy, Benchmark, and Safeguards for Elderly-Chatbot Interaction Safety
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917511851671552 |
|---|---|
| author | Fan, Changxuan Yang, Xi Zheng, Yueyuan Zhou, Bin Wang, Yuanping Hu, Wenbin Jing, Huihao Hung, Ki Sen Du, Dazhao Li, Haoran Hsiao, Janet Hui-wen Song, Yangqiu |
| author_facet | Fan, Changxuan Yang, Xi Zheng, Yueyuan Zhou, Bin Wang, Yuanping Hu, Wenbin Jing, Huihao Hung, Ki Sen Du, Dazhao Li, Haoran Hsiao, Janet Hui-wen Song, Yangqiu |
| contents | As older adults increasingly use LLM-based chatbots for companionship and assistance, a safety gap is emerging. Older adults may face vulnerabilities from social isolation, limited digital literacy, and cognitive decline, yet existing safety benchmarks largely target general harms and overlook elderly-specific risks. For example, a prompt such as "how to repair a ceiling light alone in the dark" may be benign for most users but poses a serious fall risk for older adults with mobility limitations. We introduce GrandGuard, the first comprehensive framework for assessing and mitigating elderly-specific contextual risks in LLM interactions. We develop a three-level taxonomy with 50 fine-grained risk types across mental well-being, financial, medical, toxicity, and privacy domains, grounded in real-world incidents, community discussions, and analysis of stakeholder studies. Using this taxonomy, we construct a benchmark of 10,404 labeled prompts and responses, showing that several leading LLMs mishandle elderly-specific contextual risks in over 50% of cases. We mitigate these failures with two safeguards: a fine-tuned Llama-Guard-3 and a policy-enhanced gpt-oss-safeguard-20b, achieving up to 96.2% and 90.9% unsafe-prompt detection accuracy, respectively. GrandGuard lays the groundwork for AI systems that move beyond general safety to support aging populations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_20203 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | GrandGuard: Taxonomy, Benchmark, and Safeguards for Elderly-Chatbot Interaction Safety Fan, Changxuan Yang, Xi Zheng, Yueyuan Zhou, Bin Wang, Yuanping Hu, Wenbin Jing, Huihao Hung, Ki Sen Du, Dazhao Li, Haoran Hsiao, Janet Hui-wen Song, Yangqiu Human-Computer Interaction Artificial Intelligence As older adults increasingly use LLM-based chatbots for companionship and assistance, a safety gap is emerging. Older adults may face vulnerabilities from social isolation, limited digital literacy, and cognitive decline, yet existing safety benchmarks largely target general harms and overlook elderly-specific risks. For example, a prompt such as "how to repair a ceiling light alone in the dark" may be benign for most users but poses a serious fall risk for older adults with mobility limitations. We introduce GrandGuard, the first comprehensive framework for assessing and mitigating elderly-specific contextual risks in LLM interactions. We develop a three-level taxonomy with 50 fine-grained risk types across mental well-being, financial, medical, toxicity, and privacy domains, grounded in real-world incidents, community discussions, and analysis of stakeholder studies. Using this taxonomy, we construct a benchmark of 10,404 labeled prompts and responses, showing that several leading LLMs mishandle elderly-specific contextual risks in over 50% of cases. We mitigate these failures with two safeguards: a fine-tuned Llama-Guard-3 and a policy-enhanced gpt-oss-safeguard-20b, achieving up to 96.2% and 90.9% unsafe-prompt detection accuracy, respectively. GrandGuard lays the groundwork for AI systems that move beyond general safety to support aging populations. |
| title | GrandGuard: Taxonomy, Benchmark, and Safeguards for Elderly-Chatbot Interaction Safety |
| topic | Human-Computer Interaction Artificial Intelligence |
| url | https://arxiv.org/abs/2605.20203 |