Skip to content

Risk-and-mitigation task tables: each LLM-supported task pairs a baseline prompt with task-specific risks and mitigation strategies

Design · All designs
Evidence · 1 claim (1 for) · 1 study (1 theoretical), q1 · 0 of 1 report an effect size · 1 claim rests on one study

Description

The taxonomy is instantiated as four tables (Tables 2-5), one per workflow stage, where each task row gives a definition, a model-agnostic baseline prompt, and a risk-and-mitigation column. The paper says the prompts and risk categories are designed to be transferable across tools and versions, and that "each task in the taxonomy is situated within a risk-based ethical framework and articulated through a baseline prompt." For example, summarisation carries the risk of missing key elements or introducing inaccuracies, mitigated by human reading of the document and cross-checking consistency using multiple models.

Design Implications

Context

Requirements

  • Prompts are intentionally simple and model-agnostic so students can translate journalistic intentions into explicit instructions; recurring tasks such as generation recur across stages but with different risks per editorial object

Constraints

  • The paper notes risks do not disappear but manifest differently depending on how a task is defined

Target Learners

  • journalism students

Learning Goals

  • structured prompting practice
  • ethical risk identification and mitigation
  • human oversight of AI-assisted workflows

Claims

Examples

-

Key Sources

  • Laurence Dierickx, Fredrik Bjerknes, Andreas L. Opdahl, Carl-Gustav Lindén. (2026). A Taxonomy of LLM-Supported Tasks for Critical AI Literacy in Journalism. Postdigital Science and Education. https://doi.org/10.1007/s42438-026-00690-0
Page metadata
Field Value
type design
id risk-mitigation-prompt-tables-per-stage
title Risk-and-mitigation task tables: each LLM-supported task pairs a baseline prompt with task-specific risks and mitigation strategies
description The taxonomy is instantiated as four tables (Tables 2-5), one per workflow stage, where each task row gives a definition, a model-agnostic baseline prompt, and a risk-and-mitigation column.
status review
generated by process:settle-candidates on 2026-10-10
trust unverified — no one has confirmed this page's content
sources 1 — laurence-dierickx-2026

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