Chat GPT… Some refer it is the beginning of the end. Others that it is the future happening now. Maybe somewhere in between? Each day new uses and applications of this tool are presented, imagination seems to be the limit. What about using it in cattle management and practice?
GPT stands for Generative Pre-Trained Transformer, basically it is a language model created and trained by the company OpenAI to produce optimized text through dialogue. In practice, it processes human demonstrations and compares preferences leading to adaptations according to user behaviour1.
This means that asking ChatGPT to say one phrase about how it can be helpful to a ruminant practitioner has a different result as a first question (A), or after a two-hour conversation about the subject (B), as we can see in the image below.
Example of how Chat GPT change its answer after chatting with the user.
Large language models (LLM) have the capacity to write consistent, cohesive, relevant and eloquent answers for multiple natural language processing tasks, and ChatGPT is, now, one of the models, with free access to the public, with better results, not only in translation, but also, discourse awareness and modelling abilities.
ChatGPT is a facilitator of discussions providing a more integrated point of view, since it can be used by all, and those interactions will influence in a positive way the validity of answers it will give in the future to the general public. There’re examples of positive impact of ChatGPT in the scientific and public health community, being used in class for students, to help write scientific articles and especially, specific calls for papers about the subject3.
From a way to get information on public health issues for example, it’s a tool that can relate different subjects, providing help to the population, but at the same time giving information to health workers4, using it for research and help with decision making, since it can extract and analyze data from different sources in a faster way5. A unique example of this is imagiology, where it can help directly getting data, analyze it and comparing, creating a “concept of AI-augmented radiologists”6. A similar parallel can be made for animal practitioners.
Since 2016 that the number of studies about AI and animal farming has been increasing, and more than one third of those were about cattle7. ChatGPT opens a way to communicate with producers, since its able to produce non-scientific text, that is aware of age group, social, geographic and even economic differences, adapting that text, but still transmitting valuable health information8. Working as a searching engine, it can give unified answers about multidisciplinary subjects9: new vaccinations protocols from technical governmental reports; knowledge about toxic plants in concordance to climate, geographic position and symptoms; new procedures in animal reproduction that had good economic results. In fact, the idea of an open access to both social media and medical data, can create a unique way to detect and control in very early stages not only infectious diseases, but from foodborne diseases for example10, helping in a unique way with traceability in the food chain, integrating more information “from farm to fork”11.
In a future where all the informations of a farm are connected, and learning by it self, the future becomes limitless.
There is lack of consensus between specialists about the ethical uses of LLMs. The risk of spreading misinformation and plagiarism rises questions of the necessity of rules, taxonomy and criteria of when it’s use is allowed, and to create the responsibility of admitting when a text is created, or adapted from a tool like ChatGPT, allowing moderation and filtering of scientific data associated with its use12. Some can even go further and suggest a suspension of AI in areas like healthcare till the existence of strict regulations13. A suggestion in different articles12-14, that can work around these flaws in the medical area is the existence of Open AI/LLM specifically trained and used for healthcare. Although it can limit its potential, it allows for a more trustworthy model.
Chat GPT is a powerful tool that can be used in all areas, and animal farming is surely one of them. But, as all tools, not all is beauty, there is always a potential beast. The user is a key player: asking the best questions, will obtain the best answers, and of course it will depend on the level of information that the user knows and/or needs.
In its current form, it can be used as a great and very fast search engine, giving direct answers and providing quick access to available info. But in (a probably not so very far) future, in its advanced upgraded version, it will be a very powerful tool with high potential to assist the practitioner in the decision making process.
"As a ruminant practitioner, you can find several uses for ChatGPT in your profession. Here are a few examples:
However, it's important to note that while ChatGPT can provide useful information and generate suggestions, it should not replace the expertise and judgment of a qualified ruminant practitioner. Always rely on your clinical experience and consult with colleagues or specialists when needed."
References
[1] N. Staudacher, What is ChatGPT?, (2023). https://help.openai.com/en/articles/6783457-what-is-chatgpt (accessed May 9, 2023)
[2] L. Wang, C. Lyu, T. Ji, Z. Zhang, D. Yu, S. Shi, Z. Tu, Document-Level Machine Translation with Large Language Models, ArXiv Preprint ArXiv:2304.02210. (2023). http://arxiv.org/abs/2304.02210.
[3] G. Eysenbach, The Role of ChatGPT, Generative Language Models, and Artificial Intelligence in Medical Education: A Conversation With ChatGPT and a Call for Papers, (2023). https://www.doi.org/10.2196/preprints.46885.
[4] S.S. Biswas, Role of Chat GPT in Public Health, Ann Biomed Eng. 51 (2023) 868–869. https://www.doi.org/10.1007/s10439-023-03172-7
[5] O. Baclic, M. Tunis, K. Young, C. Doan, H. Swerdfeger, Challenges and opportunities for public health made possible by advances in natural language processing, Canada Communicable Disease Report. 46 (2020) 161–168. https://www.doi.org/10.14745/ccdr.v46i06a02.
[6] T. Boeken, J. Feydy, A. Lecler, P. Soyer, A. Feydy, M. Barat, L. Duron, Artificial intelligence in diagnostic and interventional radiology: Where are we now?, Diagn Interv Imaging. 104 (2023) 1–5. https://www.doi.org/10.1016/j.diii.2022.11.004.
[7] J. Bao, Q. Xie, Artificial intelligence in animal farming: A systematic literature review, J Clean Prod. 331 (2022). https://www.doi.org/10.1016/j.jclepro.2021.129956.
[8] R. Ciappelloni, M.L. Marenzoni, S. Costarelli, Caso di studio sull’applicazione dell’Intelligenza Artificiale (AI) alla Medicina Narrativa per la scrittura di un breve racconto sulla Leptospirosi, Sanità Pubblica Veterinaria. 136 (2023).
[9] S. Neethirajan, The Use of Artificial Intelligence in Assessing Affective States in Livestock, Front Vet Sci. 8 (2021). https://www.doi.org/10.3389/fvets.2021.715261.
[10] T. Panch, J. Pearson-Stuttard, F. Greaves, R. Atun, Artificial intelligence: opportunities and risks for public health, Lancet Digit Health. 1 (2019) e13–e14. https://www.doi.org/10.1016/S2589-7500(19)30002-0
[11] A. Iftekhar, X. Cui, M. Hassan, W. Afzal, Application of Blockchain and Internet of Things to Ensure Tamper-Proof Data Availability for Food Safety, J Food Qual. 2020 (2020). https://www.doi.org/10.1155/2020/5385207
[12] J. Li, A. Dada, J. Kleesiek, J. Egger, ChatGPT in Healthcare: A Taxonomy and Systematic Review, Institute of Artificial Intelligence in Medicine, 2023. https://www.doi.org/10.1101/2023.03.30.23287899.
[13] M. Sonntagbauer, M. Haar, S. Kluge, Künstliche Intelligenz: Wie werden ChatGPT und andere KI-Anwendungen unseren ärztlichen Alltag verändern?, Med Klin Intensivmed Notfmed. (2023). https://www.doi.org/10.1007/s00063-023-01019-6.
[14] M. Hasnain, ChatGPT Applications and Challenges in Controlling Monkey Pox in Pakistan, Ann Biomed Eng. (2023). https://www.doi.org/10.1007/s10439-023-03231-z.