Parenting Journal · Volume 2 · Issue 5 · Licensed source edition
“ChatGPT, Can You Help Me Save My Child’s Life?” - Diagnostic Accuracy and Supportive Capabilities to Lay Rescuers by ChatGPT in Prehospital Basic Life Support and Paediatric Advanced Life Support Cases – An In-silico Analysis
Originally published 2023-11-21 · Journal of Medical Systems
Springer Science and Business Media LLC · DOI 10.1007/s10916-023-02019-x ↗
AI / system type
Generative AI/LLM emergency-advice safety benchmark
Family implicationsDesign, ethics, stakeholder or technical evidence with child/family implications; household benefits are not assumed.
The AI connection
The two early ChatGPT/GPT-4 systems generated diagnosis and action advice for simulated lay rescuers.
Why this reading matters
Centers a consequential trust failure: recognizing a condition and directing safe action are separate capabilities.
Editorial reading guide · Trust
Could an AI name the problem correctly while giving my family unsafe next steps?
High diagnostic recognition did not ensure safe instructions: consistent emergency-service activation occurred for only 12 of 22 scenarios, and some answers inappropriately suggested advanced procedures to parents. This distinction matters when judging urgent-sounding AI advice.
Study context
Investigators presented 22 physician-developed scenarios three times to each of ChatGPT and GPT-4, producing 132 case responses without human patients or parent participants.
Age / stage: Twenty pediatric emergency and two basic-life-support vignettes
Read with three questions
- How did diagnostic accuracy differ from safe emergency action?
- What changed when highly diagnostic wording was removed from a vignette?
- Why do local emergency resources limit transfer of these results?
Keep the limits in view
The prototypical English vignettes and zero-shot prompts were not real parent conversations. There was no human comparison group, only six responses per scenario, and safety judgments used European expectations that may not transfer to all settings.
The text below is the original article’s licensed repository edition. Its wording is not rewritten. Use the Paper Agent beside the text to locate evidence; open the source for figures, tables and complete context.
Opening the verified paper and agent…
Checking the article identity, license record, and saved source text.
Open the repository recordAttribution & reuse
Follow the original.
Original authorship, publication date, journal and DOI are preserved. This issue’s TRICK placement and reading questions are editorial additions, not claims made by the original authors.
Stefan Bushuven; Michael Bentele; Stefanie Bentele; Bianka Gerber; Joachim Bansbach; Julian Ganter; Milena Trifunovic-Koenig; Robert Ranisch (2023). “ChatGPT, Can You Help Me Save My Child’s Life?” - Diagnostic Accuracy and Supportive Capabilities to Lay Rescuers by ChatGPT in Prehospital Basic Life Support and Paediatric Advanced Life Support Cases – An In-silico Analysis Journal of Medical Systems. https://doi.org/10.1007/s10916-023-02019-x
Permissions checked against the saved article notice; repository publication notices screened 2026-11-04. This is not a new peer-review decision. The source authors and publishers are not represented as endorsing this issue.