Skip to content

Multi-layered safety architecture for teacher-supervised educational chatbots

Design · All designs
Evidence · 1 claim (1 for) · 1 study (1 qualitative), q2 · 0 of 1 report an effect size · 1 claim rests on one study

Description

Teachers in the study articulated a safety architecture for LLM-powered chatbots composed of four component types: "content scoping, presentation strategies, oversight mechanisms, and adaptation frameworks". This architecture gives teachers a reusable lesson/unit-level design that keeps chatbot behavior within professionally supervised bounds. It operationalizes the bounded authority model by layering protective mechanisms rather than relying on any single safeguard.

Design Implications

Context

Requirements

  • Human oversight mechanisms must be retained in the design

Constraints

  • Articulated by secondary teachers for LLM-powered chatbots in educational settings

Target Learners

  • secondary school students

Learning Goals

  • safe, scope-limited interaction with AI chatbots during learning

Claims

Examples

-

Key Sources

  • Reichert, H., Briceno, D., Tabarsi, B., Barnes, T. (2026). Human-Centered Design of LLM-Powered Educational Chatbots: A Study with Secondary Teachers. In: Degen, H., Ntoa, S. (eds) Artificial Intelligence in HCI. HCII 2026. LNCS vol 16746. Springer, Cham. https://doi.org/10.1007/978-3-032-31048-4_32
Page metadata
Field Value
type design
id multi-layered-safety-architecture-educational-chatbots
title Multi-layered safety architecture for teacher-supervised educational chatbots
description Teachers in the study articulated a safety architecture for LLM-powered chatbots composed of four component types: "content scoping, presentation strategies, oversight mechanisms, and adaptation frameworks".
status review
generated by process:settle-candidates on 2026-10-16
trust unverified — no one has confirmed this page's content
sources 1 — reichert-2026

Frontmatter is the page's machine-readable record — see the Schema & Guide.