Hostname: page-component-76d6cb85b7-pn7tm Total loading time: 0 Render date: 2026-07-18T03:12:56.795Z Has data issue: false hasContentIssue false
  • English
  • Français

Large Language Model (LLM)-Powered Chatbots Fail to Generate Guideline-Consistent Content on Resuscitation and May Provide Potentially Harmful Advice

Published online by Cambridge University Press:  06 November 2023

Alexei A. Birkun Open the ORCID record for Alexei A. Birkun [Opens in a new window]  and
Adhish Gautam
Alexei A. Birkun*
Affiliation:
Department of General Surgery, Anaesthesiology, Resuscitation and Emergency Medicine, Medical Academy named after S.I. Georgievsky of V.I. Vernadsky Crimean Federal University, Simferopol, 295051, Russian Federation
Adhish Gautam
Affiliation:
Regional Government Hospital, Una (H.P.), 174303, India
*
Correspondence: Alexei A. Birkun, MD, DMedSc Medical Academy named after S.I. Georgievsky of V.I. Vernadsky Crimean Federal University Lenin Blvd, 5/7, Simferopol, 295051, Russian Federation E-mail: birkunalexei@gmail.com
Get access Rights & Permissions [Opens in a new window]

Abstract

Introduction:

Innovative large language model (LLM)-powered chatbots, which are extremely popular nowadays, represent potential sources of information on resuscitation for the general public. For instance, the chatbot-generated advice could be used for purposes of community resuscitation education or for just-in-time informational support of untrained lay rescuers in a real-life emergency.

Study Objective:

This study focused on assessing performance of two prominent LLM-based chatbots, particularly in terms of quality of the chatbot-generated advice on how to give help to a non-breathing victim.

Methods:

In May 2023, the new Bing (Microsoft Corporation, USA) and Bard (Google LLC, USA) chatbots were inquired (n = 20 each): “What to do if someone is not breathing?” Content of the chatbots’ responses was evaluated for compliance with the 2021 Resuscitation Council United Kingdom guidelines using a pre-developed checklist.

Results:

Both chatbots provided context-dependent textual responses to the query. However, coverage of the guideline-consistent instructions on help to a non-breathing victim within the responses was poor: mean percentage of the responses completely satisfying the checklist criteria was 9.5% for Bing and 11.4% for Bard (P >.05). Essential elements of the bystander action, including early start and uninterrupted performance of chest compressions with adequate depth, rate, and chest recoil, as well as request for and use of an automated external defibrillator (AED), were missing as a rule. Moreover, 55.0% of Bard’s responses contained plausible sounding, but nonsensical guidance, called artificial hallucinations, that create risk for inadequate care and harm to a victim.

Conclusion:

The LLM-powered chatbots’ advice on help to a non-breathing victim omits essential details of resuscitation technique and occasionally contains deceptive, potentially harmful directives. Further research and regulatory measures are required to mitigate risks related to the chatbot-generated misinformation of public on resuscitation.

Keywords

artificial hallucinationartificial intelligencecardiac arrestcardiopulmonary resuscitationchatbotlarge language model

Information

Type
Original Research
Information
Prehospital and Disaster Medicine , Volume 38 , Issue 6 , December 2023 , pp. 757 - 763
Check for updates
Copyright
© The Author(s), 2023. Published by Cambridge University Press on behalf of the World Association for Disaster and Emergency Medicine

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Article purchase

Temporarily unavailable

References

Haleem, A, Javaid, M, Singh, RP. An era of ChatGPT as a significant futuristic support tool: a study on features, abilities, and challenges. BenchCouncil Transactions on Benchmarks, Standards, and Evaluations. 2022;2(4):100089.CrossRefGoogle Scholar
De Angelis, L, Baglivo, F, Arzilli, G, et al. ChatGPT and the rise of large language models: the new AI-driven infodemic threat in public health. Front Public Health. 2023;11:1166120.CrossRefGoogle ScholarPubMed
Hassani, H, Silva, ES. The role of ChatGPT in data science: how AI-assisted conversational interfaces are revolutionizing the field. Big Data Cogn Comput. 2023;7(2):62.CrossRefGoogle Scholar
Ahn, C. Exploring ChatGPT for information of cardiopulmonary resuscitation. Resuscitation. 2023;185:109729.CrossRefGoogle ScholarPubMed
Altamimi, I, Altamimi, A, Alhumimidi, AS, Altamimi, A, Temsah, MH. Snakebite advice and counseling from artificial intelligence: an acute venomous snakebite consultation with ChatGPT. Cureus. 2023;15(6):e40351.Google ScholarPubMed
Birkun, AA, Gautam, A. Instructional support on first aid in choking by an artificial intelligence-powered chatbot. Am J Emerg Med. 2023;70:200–202.CrossRefGoogle ScholarPubMed
Dahdah, JE, Kassab, J, Helou, MCE, Gaballa, A, Sayles, S 3rd, Phelan, MP. ChatGPT: a valuable tool for emergency medical assistance. Ann Emerg Med. 2023;82(3):411–413.CrossRefGoogle ScholarPubMed
Fijačko, N, Gosak, L, Štiglic, G, Picard, CT, John Douma, M. Can ChatGPT pass the life support exams without entering the American Heart Association course? Resuscitation. 2023;185:109732.CrossRefGoogle ScholarPubMed
Sarbay, İ, Berikol, GB, Özturan, İU. Performance of emergency triage prediction of an open access natural language processing based chatbot application (ChatGPT): a preliminary, scenario-based cross-sectional study. Turkish J Emerg Med. 2023;23(3):156.CrossRefGoogle ScholarPubMed
Berg, KM, Cheng, A, Panchal, AR, et al. Part 7: Systems of Care: 2020 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation. 2020;142(16_suppl_2):S580–S604.CrossRefGoogle ScholarPubMed
Semeraro, F, Greif, R, Böttiger, BW, et al. European Resuscitation Council Guidelines 2021: systems saving lives. Resuscitation. 2021;161:80–97.CrossRefGoogle ScholarPubMed
Liu, KY, Haukoos, JS, Sasson, C. Availability and quality of cardiopulmonary resuscitation information for Spanish-speaking population on the Internet. Resuscitation. 2014;85(1):131–137.CrossRefGoogle ScholarPubMed
Metelmann, B, Metelmann, C, Schuffert, L, Hahnenkamp, K, Brinkrolf, P. Medical correctness and user friendliness of available apps for cardiopulmonary resuscitation: systematic search combined with guideline adherence and usability evaluation. JMIR Mhealth Uhealth. 2018;6:e190.CrossRefGoogle ScholarPubMed
Birkun, A, Gautam, A, Trunkwala, F, Böttiger, BW. Open online courses on basic life support: availability and resuscitation guidelines compliance. Am J Emerg Med. 2022;62:102–107.CrossRefGoogle ScholarPubMed
Birkun, AA, Dr, Gautam A.. Google’s advice on first aid: evaluation of the search engine’s question-answering system responses to queries seeking help in health emergencies. Prehosp Disaster Med. 2023;38(3):345–351.CrossRefGoogle ScholarPubMed
Perkins, GD, Colquhoun, M, Deakin, CD, et al. Resuscitation Council UK. 2021 Resuscitation Guidelines. Adult Basic Life Support Guidelines, 2021. https://www.resus.org.uk/library/2021-resuscitation-guidelines/adult-basic-life-support-guidelines. Accessed August 15, 2023.Google Scholar
Birkun, A, Gautam, A. Dataset of analysis of the large language model-powered chatbots’ advice on help to a non-breathing victim. Mendeley Data. 2023;V1.CrossRefGoogle Scholar
Kincaid, JP, Fishburne, RP Jr, Rogers, RL, Chissom, BS. Derivation of New Readability Formulas (Automated Readability Index, Fog Count, and Flesch Reading Ease Formula) for Navy Enlisted Personnel. Millington, Tennessee USA: Naval Technical Training Command, Millington TN Research Branch; 1975.CrossRefGoogle Scholar
Datayze. Readability Analyzer. https://datayze.com/readability-analyzer. Accessed August 15, 2023.Google Scholar
Peters, J. The Bing AI bot has been secretly running GPT-4. The Verge. https://www.theverge.com/2023/3/14/23639928/microsoft-bing-chatbot-ai-gpt-4-llm. Accessed August 15, 2023.Google Scholar
Microsoft Bing. Bing Webmaster Guidelines. https://www.bing.com/webmasters/help/webmasters-guidelines-30fba23a. Accessed August 15, 2023.Google Scholar
Bard. Bard FAQ. https://bard.google.com/faq. Accessed August 15, 2023.Google Scholar
Search Engine Land. SEO. Breaking Bard: Google’s AI chatbot lacks sources, hallucinates, gives bad SEO advice. https://searchengineland.com/google-bard-first-looks-394583. Accessed August 15, 2023.Google Scholar
Alkaissi, H, McFarlane, SI. Artificial hallucinations in ChatGPT: implications in scientific writing. Cureus. 2023;15:e35179.Google ScholarPubMed
Bickmore, TW, Trinh, H, Olafsson, S, et al. Patient and consumer safety risks when using conversational assistants for medical information: an observational study of Siri, Alexa, and Google Assistant. J Med Internet Res. 2018;20(9):e11510.CrossRefGoogle ScholarPubMed
Picard, C, Smith, KE, Picard, K, Can Alexa, Douma MJ., Cortana, Google Assistant and Siri save your life? A mixed-methods analysis of virtual digital assistants and their responses to first aid and basic life support queries. BMJ Innovations. 2020;6.CrossRefGoogle Scholar
Bing. Introducing the new Bing. https://www.bing.com/new. Accessed August 15, 2023.Google Scholar
Supplementary material: File

Birkun and Gautam supplementary material

Birkun and Gautam supplementary material

Download Birkun and Gautam supplementary material(File)
File 12.9 KB