As the world prepares for the possibility of another global pandemic, the integration of artificial intelligence (AI) into health systems could play a crucial role in the next outbreak. Often referred to as "Disease X", the next potential pandemic is a looming threat, with some experts predicting a one in four chance of another outbreak as severe as COVID-19 within the next decade. While the exact nature of the disease could be influenza, coronavirus, or something entirely new, the devastation caused by COVID-19 has led to a rethinking of global preparedness. Central to these discussions is whether AI can significantly aid in preventing or mitigating the effects of the next pandemic.
AI's Role in Early Detection
Researchers in California are already working on AI-based systems designed to provide early warnings for potential pandemics. At the forefront of this research are teams from the University of California, Irvine (UCI) and the University of California, Los Angeles (UCLA), who are part of the US National Science Foundation's Predictive Intelligence for Pandemic Prevention program. This initiative focuses on utilizing AI to predict, track, and mitigate the effects of future pandemics by analyzing large datasets.
One of their key projects involves the development of a tool that monitors social media platforms, such as Twitter (now known as X). By analyzing billions of tweets, researchers aim to identify early warning signs of possible outbreaks. Professor Chen Li, leading the project at UCI, explains that their AI model can scan and categorize relevant social media posts to detect trends indicative of an impending epidemic. Additionally, the tool could help public health departments predict outbreaks and assess the impact of different public health policies. However, challenges remain, particularly with data availability in non-US regions where access to platforms like X may be restricted.
AI in Mutation Prediction and Vaccine Development
Beyond early detection, AI is also being utilized to predict the mutations of viruses and help pharmaceutical companies expedite vaccine development. EVEScape, an AI tool developed by Harvard Medical School and the University of Oxford, ranks new variants of the coronavirus every two weeks and has also made accurate predictions for viruses like HIV and influenza. According to Nikki Thadani, a former postdoctoral research fellow involved in the project, EVEScape's predictive abilities are especially valuable early in a pandemic, helping vaccine manufacturers and researchers anticipate potential mutations and design better therapeutics.
AstraZeneca, a global pharmaceutical giant, is another key player using AI to accelerate antibody discovery. According to Jim Wetherall, AstraZeneca's vice president of data science and AI R&D, AI has dramatically shortened the timeline for identifying effective antibody candidates—from three months to just three days. By generating and screening a vast library of antibodies, AI enables faster responses to rapidly mutating viruses, which is critical in fast-moving pandemics.
AI's Broader Role in Pandemic Preparedness
The Coalition for Epidemic Preparedness Innovations (CEPI), headquartered in Oslo, has also turned to AI in its efforts to prepare for and respond to future epidemics. CEPI’s aim is to develop vaccines for emerging infectious diseases, and AI plays a key role in speeding up this process. Dr. In-Kyu Yoon, CEPI’s director of innovative technology, highlights that AI accelerates various aspects of preparation, from vaccine development to real-time response analysis.
However, while AI offers tremendous potential, it is not without its limitations. As Dr. Yoon points out, AI's predictions are only as good as the data it is trained on. Despite its impressive capabilities, AI still relies heavily on human input and decision-making. Without high-quality, comprehensive data, its effectiveness can be compromised. Therefore, while AI can significantly enhance preparation efforts, it cannot prevent or slow down a pandemic on its own. It remains a tool that must be strategically applied by humans.
Challenges and Limitations
AI’s potential to transform pandemic preparedness is immense, but it comes with challenges. The reliance on platforms like X (formerly Twitter) for data, especially in regions where these platforms are restricted, presents an obstacle. Moreover, the scarcity of data outside the US introduces potential biases when attempting to apply predictive models globally.
Additionally, while AI has shown remarkable speed and efficiency in certain areas, such as antibody discovery and vaccine development, it is important to acknowledge that AI is still evolving. As noted by several experts, including those at CEPI and AstraZeneca, AI's effectiveness is limited by the quality and completeness of the data it analyzes. It cannot replace human oversight, as human judgment is crucial in determining how and where to apply AI's capabilities effectively.
In conclusion, AI represents a powerful tool in the fight against future pandemics, offering faster detection, prediction of virus mutations, and accelerated vaccine development. However, it is not a magic bullet. Its success depends on the quality of the data it processes and the expertise of the humans guiding its use. As the world braces for the next "Disease X," AI will undoubtedly be part of the solution, but it will be up to people to ensure its proper application in preventing and mitigating the impacts of future pandemics.
