Answers are grounded in this paper and cite their sources.
Adolescent mental health challenges are worsening. Traditional paper- or web-based screening tools lack generalizability, timeliness, and objectivity—and are especially vulnerable to social desirability bias and stigma, causing students to conceal distress. AI-generated content (AIGC) offers a promising alternative for mental health assessment. Yet, the validity and acceptability of AIGC-based conversational assessments in real school settings remain untested. This study involved 584 seventh- and eighth-grade students from Chunjiang Middle School (Hangzhou, Zhejiang) to: (1) assess agreement between the AIGC conversational model and the school's existing observation list; (2) test its ability to identify at-risk students missed by routine observation; and (3) evaluate student acceptance. Results show the AIGC model aligns well with current practices, detects overlooked cases, and is well accepted—making it a viable supplement to traditional screening and teacher observation. It supports a shift from passive intervention to active prevention in school mental health. This study provides empirical evidence and practical guidance for scaling AIGC-based screening in K–12 settings and advances efforts to build a destigmatized adolescent mental health service system.
No takes yet. Share an insight, caveat, or question.
Zhu Dingchen (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: