Risk category explains the largest share of variation in educational attack success, far more than LLM usage context or curriculum topic¶
Claim · All claims
Evidence · 1 study · designr2·q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person · recheck by 2029: tool and model accuracy changes quickly
Subclaims¶
Reading the evidence codes
q is the design tier, i the impact, n the sample, and the kind with its r
the rigour judged within that kind: what each code means.
This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.
q2 i? In a binomial GLM of conversation-level attack success, risk category accounts for 77.4% of explained deviance and interaction setting 22.9%, while usage context (7.7%) and curriculum topic (0.9%) contribute relatively little. → Park 2026
Evidence¶
Park 2026¶
Park, J., Han, J., Yoo, H., Ahn, S.-Y., Yoon, J., & Oh, A. (2026). EduZone: A Framework for Evaluating LLM Safety for K-12 Students and Teachers. arXiv preprint. https://arxiv.org/abs/2608.02024 doi:10.48550/arxiv.2608.02024
q2 · i? · design · r2
Binomial GLM fitted with conversation-level attack success as the binary outcome to identify main factors influencing educational safety. The authors focus subsequent analysis on risk category and interaction setting given the other factors' small shares.
"Risk category accounts for the largest share of the explained deviance (77.4%), followed by interaction setting (22.9%)"
Discussion¶
Related Claims¶
- LLM attack success rates increase consistently from single-turn to static multi-turn and dynamic multi-turn interactions in educational settings — related
- Current LLMs are more vulnerable to education-specific risks such as academic misconduct and excessive cognitive load than to conventional safety risks in K-12 educational interactions — related
- No evaluated LLM tutor is reliably safe: every model exceeds 60% harm rate on at least five risk categories in single-turn and six in multi-turn evaluation — related
- Task category is the dominant factor in agent performance: cross-category spread (26.3 points) far exceeds cross-model spread (8.8 points) — related
- Adaptive multi-turn interactions substantially amplify LLM vulnerabilities to conventional safety risks in educational contexts — related
- Taxonomy-augmented guardrail classifiers reduce educational attack success more than general-purpose jailbreak defenses — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
risk-category-dominates-educational-safety-variance |
title |
Risk category explains the largest share of variation in educational attack success, far more than LLM usage context or curriculum topic |
description |
Risk category explains the largest share of variation in educational attack success, far more than LLM usage context or curriculum topic |
status |
review |
evidence_strength |
moderate |
newest_evidence |
2026 |
recheck_by |
2029 |
generated |
by process:wiki-ingest on 2026-10-10 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — park-2026 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.