Safe-Child-LLM: A Developmental Benchmark for Evaluating LLM Safety in Child-LLM Interactions

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Jiao, Junfeng, Afroogh, Saleh, Chen, Kevin, Murali, Abhejay, Atkinson, David, Dhurandhar, Amit
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914593044955136
author Jiao, Junfeng
Afroogh, Saleh
Chen, Kevin
Murali, Abhejay
Atkinson, David
Dhurandhar, Amit
author_facet Jiao, Junfeng
Afroogh, Saleh
Chen, Kevin
Murali, Abhejay
Atkinson, David
Dhurandhar, Amit
contents As Large Language Models (LLMs) increasingly power applications used by children and adolescents, ensuring safe and age-appropriate interactions has become an urgent ethical imperative. Despite progress in AI safety, current evaluations predominantly focus on adults, neglecting the unique vulnerabilities of minors engaging with generative AI. We introduce Safe-Child-LLM, a comprehensive benchmark and dataset for systematically assessing LLM safety across two developmental stages: children (7-12) and adolescents (13-17). Our framework includes a novel multi-part dataset of 200 adversarial prompts, curated from red-teaming corpora (e.g., SG-Bench, HarmBench), with human-annotated labels for jailbreak success and a standardized 0-5 ethical refusal scale. Evaluating leading LLMs -- including ChatGPT, Claude, Gemini, LLaMA, DeepSeek, Grok, Vicuna, and Mistral -- we uncover critical safety deficiencies in child-facing scenarios. This work highlights the need for community-driven benchmarks to protect young users in LLM interactions. To promote transparency and collaborative advancement in ethical AI development, we are publicly releasing both our benchmark datasets and evaluation codebase at https://github.com/The-Responsible-AI-Initiative/Safe_Child_LLM_Benchmark.git
format Preprint
id arxiv_https___arxiv_org_abs_2506_13510
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Safe-Child-LLM: A Developmental Benchmark for Evaluating LLM Safety in Child-LLM Interactions
Jiao, Junfeng
Afroogh, Saleh
Chen, Kevin
Murali, Abhejay
Atkinson, David
Dhurandhar, Amit
Computers and Society
As Large Language Models (LLMs) increasingly power applications used by children and adolescents, ensuring safe and age-appropriate interactions has become an urgent ethical imperative. Despite progress in AI safety, current evaluations predominantly focus on adults, neglecting the unique vulnerabilities of minors engaging with generative AI. We introduce Safe-Child-LLM, a comprehensive benchmark and dataset for systematically assessing LLM safety across two developmental stages: children (7-12) and adolescents (13-17). Our framework includes a novel multi-part dataset of 200 adversarial prompts, curated from red-teaming corpora (e.g., SG-Bench, HarmBench), with human-annotated labels for jailbreak success and a standardized 0-5 ethical refusal scale. Evaluating leading LLMs -- including ChatGPT, Claude, Gemini, LLaMA, DeepSeek, Grok, Vicuna, and Mistral -- we uncover critical safety deficiencies in child-facing scenarios. This work highlights the need for community-driven benchmarks to protect young users in LLM interactions. To promote transparency and collaborative advancement in ethical AI development, we are publicly releasing both our benchmark datasets and evaluation codebase at https://github.com/The-Responsible-AI-Initiative/Safe_Child_LLM_Benchmark.git
title Safe-Child-LLM: A Developmental Benchmark for Evaluating LLM Safety in Child-LLM Interactions
topic Computers and Society
url https://arxiv.org/abs/2506.13510