Feature / Essayprompt engineeringjournalismgenerative AI
Journalism guide outlines four role-playing prompt techniques to improve reporters' use of generative AI
5.5
A practical guide outlines four role-playing prompt techniques journalists can use with generative AI: rubber ducking, critical-friend/mentoring, red teaming/devil's-advocate, and audience personas. The piece supplies example prompts and templates and recommends grounding prompts with documents and using negative prompting to constrain model 'greediness.' It warns of risks such as cognitive offloading while arguing role play is a low-risk, high-value way to improve reporting rigor and editorial workflows.
Key Points
- 1Core technical detail: Four prompt patterns (rubber ducking, mentor/critical friend, red team/devil's advocate, audience personas) plus negative prompting and grounding to mitigate over-compliance and hallucination.
- 2Business implication: Newsrooms can use these structured prompts to increase editorial rigor, accelerate verification, and train journalists without exposing audiences to inaccurate AI-generated outputs.
- 3Future impact: Widespread adoption could shift newsroom workflows toward AI-assisted iterative critique and testing, while raising training needs to avoid cognitive offloading and preserve journalists' skills.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
