Predicting future pandemics and formulating prevention strategies: The role of ChatGPT
![Closeup hand with magnifying glass over the smartphone](/sites/default/files/styles/landscape_16_9/public/2022-12/Shutterstock_1840565794-min.jpg?h=2cf907fb&itok=mCQFyh_z)
Abstract
The emergence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in late 2019 and its subsequent worldwide spread has emphasized the urgent need for better approaches for predicting and managing infectious disease outbreaks. One potential instrument in this effort is artificial intelligence (AI), and in specific, language models like ChatGPT (Chat Generative Pre-trained Transformer). In the present study, to explore how ChatGPT could predict future pandemics and give suggestions about the prevention strategy, our research team chatted with ChatGPT with several questions on July 12, 2023. Based on our conversation, we can conclude that AI is not a substitute for human expertise, but an adjunct to support early prediction, prevention, and management of future pandemics.
Editorial
Since the World Health Organization announced that coronavirus disease 2019 (COVID-19) is no longer a public health emergency of international concern [1], a key question has started arising: Which organism will most likely cause the next pandemic? Throughout the centuries, the world has been afflicted by a range of infectious diseases such as the bubonic plague, Spanish flu, severe acute respiratory syndrome (SARS), Middle East respiratory syndrome (MERS), Ebola, Nipah, and more recently, COVID-19. The rapid globalization and increased travel across continents in modern times have further amplified the capacity of these diseases to evolve as pandemics, creating a persistent threat to human survival. The golden lesson we learned from the recent COVID-19 pandemic is that prevention beats cure. Despite substantial progress in scientific and technological advancements, the COVID-19 pandemic faced various challenges, such as a lack of sufficient prior knowledge of biological threats and public health preparedness. In this context, an innovative approach for predicting future pandemics is the key factor in boosting efficiency and preparedness for future encounters. Predicting future events with conviction is beyond our grip. However, by considering past outbreaks, analyzing large datasets, and assessing multiple factors and variables, we can gather hints about what pandemic may lie ahead. The first attempt to predict the spread of disease was made in the 14th century during the historical Black Death epidemic in Europe [2]. In 1850, during a cholera outbreak in London, John Snow analyzed the available data and proposed that certain areas served by a particular water pump were more affected than others [3]. Over the last few months, researchers explored the utility of machine learning approaches to predict the COVID-19 outbreak. Sina F. Ardabili et al. deployed a machine learning model to predict the COVID-19 outbreak as a substitute for susceptible-exposed-infectious-removed (SEIR) and susceptible-infected-recovered (SIR) models and they suggested machine learning as an effective tool for long-term prediction [4]. Another study by Gergo Pinter et al. demonstrated the potential of a hybrid machine-learning approach for predicting the COVID-19 pandemic in Hungary [5]. Based on the coronavirus's whole genome sequence, deep learning models have been fruitfully used in another study to predict the likelihood of cross-species infection for early pandemic warnings [6]. Investigators also explored the capability of ChatGPT (Chat Generative Pre-trained Transformer; OpenAI, San Francisco, CA) in predicting common drug-drug interactions [7], and upcoming technology on diabetes [8]. To the best of our knowledge, there is no existing literature examining the capability of this artificial intelligence (AI) tool to predict future pandemics. In this present manuscript, we will explore the role of ChatGPT in predicting future pandemics and its potential to improve our capability to manage upcoming encounters.
In the era of AI, ChatGPT is currently in a boom and already is used by millions of users. This natural language processing (NLP) model was developed by the San Francisco-based AI research and deployment organization OpenAI and released in November 2022 [9]. Its popularity has been mostly attributed to its capacity to engage in natural language conversations. The emergence of a novel organism, alternations of environmental conditions, host-pathogen interaction, and worldwide travel patterns are the various factors that directly or indirectly influence our predictive proficiencies regarding a pandemic. In this context, ChatGPT can enhance our predictive capabilities by evaluating epidemiological, environmental, clinical, news article, and social media datasets, as well as satellite imagery, health surveillance, and genomic sequencing data. Examining the data on the genetic makeup of pathogens and relating them to already existing treatments and vaccines, this AI tool can accelerate the development of potential treatments and vaccines for new pathogens. Our research team talked with ChatGPT on July 12, 2023, to explore how it might predict the future pandemic and give suggestions on prevention strategies from the AI dimensions. The entire conversation is described in Table 1.
As the conversation ended, it is important to note that guessing the future pandemic is a multidimensional and challenging assignment, and there are limitations to what AI language models like ChatGPT can accomplish. As precise prediction depends on the quality and quantity of the already generated dataset and AI models are completely dependent on these datasets, they may face biases that are incorporated in the data. To encounter this limitation, collaboration among public health experts, clinicians, researchers, and policymakers across the globe is needed to construct good quality and diverse datasets that will mimic a wide range of inhabitants and atmospheres. Protecting the privacy and security of health information is another challenge during global data sharing. In this regard, there is an urgent need to formulate strict guidelines and standard operating procedures to overcome the challenges. Another limitation of the AI model is the probability of false positive and false negative predictions. False positive prediction might create excessive anxiety and overreaction in the community, whereas false negative prediction can create delayed response leading to increased morbidity and mortality. Despite these drawbacks, ChatGPT is an appreciated model that will help in improving our capability to predict and respond to a pandemic.
In conclusion, predicting future pathogens and formulating prevention strategies in advance is crucial for a prompt and effective pandemic response. AI language models like ChatGPT are not a replacement for human knowledge, experience, and interpretation. These models might deliver valuable insights and predictions of potential outbreaks on the basis of published or available data; it is ultimately the job of public health experts and policymakers to make necessary decisions based on these predictions, experiences, and feedback from the researchers. Moreover, the application of this AI tool in the prediction model is yet to be explored further by mapping the dimensions of various challenges.