Normal View Dyslexic View

View from the coffee room…LARPing ChatGPT

Virve Koljonen MD, PhD

Department of Plastic surgery, Helsinki University and Helsinki University Hospital, Helsinki Finland

3 February 2025
Guest blog General
BJSA
BJS Academy
0000-0000
BJS Foundation Limited
London, UK
LARP, live action role play, is immersive role-playing where participants physically and mentally/emotionally portray their characters, interact in real-world settings, and shape the story through improvisation. There are not many chances for surgeons to LARP during professional hours. Well, other than occasionally LARP internist, when someone asks about medicines.
Medical research and writing have changed after introduction to LLMs, large language models, such as ChatGPT (OpenAI, San Francisco, CA, USA) launched Nov 30, 2022.  LLMs employ neural network and have been trained to understand and generate human language and  produce human-like responses1.
It has been stated that LLMs will revolutionize or maybe already have revolutionized medical research and writing2-4. Wow, strong bold words. This revolution opens a new LARP experience: LARP ChatGPT in medical writing. To fully immerse ourselves in the character of ChatGPT there are few things to need and consider when producing LARP ChatGPT texts. Let me be your guide through some of the main features, how to excel in LARP ChatGPT.
Let’s start with words; very, very important, without words, there is no article. If you don’t have words, then the even the most ingenious idea or research finding is just daydreaming. Not just any words, more specifically you want to use words that are ChatGPT-specific. After the launch of ChatGPT in November 2022, there has been an excess of certain words, compared with pre-ChatGPT in medical articles.  Such words are delves, showcasing, underscores, potential, findings, crucial5 .  In addition to these, meticulous and commendable 6 are definitely words you want to use.  I give an example of a title; you may want to edit this to make it more suitable for your area of expertise: Meticulous delving into commendable use of surgery does not underscore potential and crucial results of our findings.  See, already we are feeling like AI!
How about style? Medical writing has its own and unique style or should I say nowadays, human medical writing style. Comparing human writing style to ChatGPT style disclosed that human texts are concrete, more diverse, and typically contain more useful information7. That sounds so boring. The ChatGPT texts are fluent and use general terminology rather than information specific to the issue at hand7.  Remember to use  a lot of adjectives and adverbs7.  You definitely want to be repetitive, and use the same expressions over and over in your paper6.  One very important notion is to make your article short, preferably less that 400 words and be sure to write unrelated to the actual topic8.  As your mentor in this LARP experience I will give you again an example of good sentences; this is about emergency surgery, I made this up myself:  More early and thus also most quickly surgical surgery is in very much benefit  of [surgically treatable disease]. Beautiful and colorful surgical scrubs may also improve the mind set of attending and highly qualified OR team, that is multiprofessional for the advantage for the suffering patients.  Yes! We have adjectives, adverbs, general term for intervention - surgery, and the text flows like a spring creek, brought into being from snow kissed by the sun's first rays.
Phrases are typical in scientific medical writing. One of the examples that comes to mind is: to the best of our knowledge, meaning that we did not actually make a through literature review, just browsed few articles, and thus we are not responsible for this argument. However, to actually LARP ChatGPT you must at all costs avoid using phrases such as  percentage of, was measured, number of, with respect, respect to, and to determine 6.  Instead you want to include phrases occurrence of, these findings, have shown and, interaction between 6 in your article. I am confident that you can manage this phrases thing by now. If I may suggest something; use interaction between, it showcases crucial potential. All right, I give you an example: These findings in this current and very commendable paper, we have showninteraction between and occurrence of professional and skillful surgical  surgery and wound healing.
To really make your article stand out, I suggest that you use one or all of these sentences9: as of my last knowledge update in [month year], Certainly, here is/are [a possible introduction introduction/results/discussion for your topic/here is the text without the bullet points/here is a literature survey], I don’t have access to. Many of fellow ChatGPT LARPers are already using them9.
Scientific writing and communication are built on previous research, or as we like to call it references. I can attest that now and again, writing scientific article calls for spending 6 hours reading 35 articles so one can write one sentence containing two references, as pointed out cleverly in the popular social media meme. Now I have great news for you. In LARP ChatGPT there is no need to read for six hours, you can just invent the references10! Be sure though, to make your invented references look like legit at first sight, and only after careful examination by those pesky reviewers they appear as what they are - a  hoax10.  Hallucination, absence of factuality10  or as we human writers call it lying, should absolutely be incorporated in your text. Apart from inventing references you can actually make up whole sentences and even paragraphs. Don’t let restrictive treatment recommendations get in your way.
So, you have now mastered ChatGPT LARP, and your paper is ready. The next step is to submit it. To get a fully immersive experience of your LARP, I suggest that you submit your paper into non-indexed journal11 and preferably predator or vanity journal.  After that, you may reach the ultimate ChatGPT LARP goal: your paper is included in the list of suspected ChatGPT misuse11.
Hope this helps.
References
1Sallam M. ChatGPT utility in healthcare education, research, and practice: systematic review on the promising perspectives and valid concerns. In: Healthcare; 2023: MDPI; 2023. p. 887.
2Lee PY, Salim H, Abdullah A, Teo CH. Use of ChatGPT in medical research and scientific writing. Malaysian Family Physician: the Official Journal of the Academy of Family Physicians of Malaysia 2023;18: 58.
3Mesko B. The ChatGPT (Generative Artificial Intelligence) Revolution Has Made Artificial Intelligence Approachable for Medical Professionals. J Med Internet Res 2023;25: e48392.
4Diaz Milian R, Moreno Franco P, Freeman WD, Halamka JD. Revolution or Peril? The Controversial Role of Large Language Models in Medical Manuscript Writing. Mayo Clin Proc 2023;98(10): 1444-1448.
5Kobak D, Márquez RG, Horvát E-Á, Lause J. Delving into ChatGPT usage in academic writing through excess vocabulary. arXiv preprint arXiv:240607016 2024.
6Matsui K. Delving into PubMed Records: Some Terms in Medical Writing Have Drastically Changed after the Arrival of ChatGPT. medRxiv 2024: 2024.2005. 2014.24307373.
7Liao W, Liu Z, Dai H, Xu S, Wu Z, Zhang Y, Huang X, Zhu D, Cai H, Li Q, Liu T, Li X. Differentiating ChatGPT-Generated and Human-Written Medical Texts: Quantitative Study. JMIR Med Educ 2023;9: e48904.
8Ariyaratne S, Iyengar KP, Nischal N, Chitti Babu N, Botchu R. A comparison of ChatGPT-generated articles with human-written articles. Skeletal Radiol 2023;52(9): 1755-1758.
9Strzelecki A. ‘As of my last knowledge update’: How is content generated by ChatGPT infiltrating scientific papers published in premier journals? Learned Publishing 2025;38(1): e1650.
10Rafaqat W, Chu DI, Kaafarani HM. AI and ChatGPT Meet Surgery: A Word of Caution for Surgeon-Scientists. Ann Surg 2023;278(5): e943-e944.
11Haider J, Söderström KR, Ekström B, Rödl M. GPT-fabricated scientific papers on Google Scholar: Key features, spread, and implications for preempting evidence manipulation. Harvard Kennedy School Misinformation Review 2024;5(5).
Info
Copied!