Researchers from The Ethox Centre at Oxford Population Health have contributed to a Nature perspective paper outlining, for the first time, how artificial intelligence (AI) can revolutionize infectious disease research and enhance pandemic preparedness. Over the next five years, integrating AI into national response systems could save lives by predicting the location and trajectory of disease outbreaks. The global team of researchers emphasizes the need for stronger collaboration between academia, government, and industry to ensure AI is used safely, ethically, and with accountability in infectious disease research.
Published following last week’s AI Action Summit and amid growing global discussions on AI investment and regulation, the study highlights the importance of prioritizing safety, accountability, and ethics in deploying AI for infectious disease research.
Professor Michael Parker, Director of the Ethox Centre, stated: "This paper underscores the potential of AI in pandemic preparedness, prevention, and response. It also stresses that the success of these technologies depends on identifying and analyzing relevant ethical considerations while fostering public trust and confidence through well-grounded, sustainable approaches. Achieving this will require high-quality, collaborative ethics research, engaging scholars from low-, middle-, and high-income countries."
The study, which calls for a transparent and cooperative approach in sharing datasets and AI models, is the result of a partnership between scientists and ethicists from the University of Oxford and colleagues from academia, industry, and policy organizations across Africa, America, Asia, Australia, and Europe.
So far, AI in medicine has primarily focused on individual patient care, enhancing diagnostics, precision medicine, and clinical decision-making. This review, however, explores AI’s role in population health. The researchers find that AI is increasingly capable of delivering reliable results even with limited data, a key challenge in the field. These advancements open new possibilities for AI to improve healthcare across both high- and low-income countries.
Lead author Professor Moritz Kraemer from the University of Oxford’s Pandemic Sciences Institute said, "AI has the potential to transform pandemic preparedness in the next five years.
"It will enable us to anticipate outbreak locations and predict their progression by analyzing vast amounts of climatic and socio-economic data. AI could also help assess how emerging pathogens interact with the immune system, shedding light on their potential impact on individual patients.
"If integrated into national response systems, these innovations could save lives and significantly strengthen global preparedness for future pandemics."
Key opportunities for AI in pandemic preparedness identified in the research include:
Enhancing current disease spread models to improve accuracy and realism
Identifying high-transmission areas to optimize resource allocation
Improving genetic data analysis for faster vaccine development and variant detection
Predicting characteristics of new pathogens and their likelihood of cross-species transmission
Forecasting the emergence of new variants of circulating viruses like SARS-CoV-2 and influenza, and determining the best treatment and vaccine strategies
Integrating population-level data with wearable device data—such as heart rate and activity levels—to better detect and monitor outbreaks
Bridging the gap between complex scientific research and healthcare professionals with limited AI expertise, boosting capacity in under-resourced settings
However, not all areas of pandemic preparedness will be equally affected by AI advancements. For instance, while AI-driven protein language models could significantly accelerate understanding of virus mutations, foundational AI models may only offer modest improvements in tracking pathogen spread.
The researchers caution against the assumption that AI alone can solve infectious disease challenges. Instead, they advocate for incorporating human expertise into AI modeling workflows to overcome existing limitations.
They also highlight concerns about data quality, representation biases, restricted AI model accessibility, and risks associated with relying on opaque, “black-box” AI systems for decision-making.
Professor Eric Topol, founder and director of the Scripps Research Translational Institute, emphasized: "AI’s transformative potential for pandemic mitigation depends on global collaboration and continuous, comprehensive surveillance data."
Study lead author Samir Bhatt from the University of Copenhagen and Imperial College London added: "Infectious disease outbreaks remain a constant threat, but AI provides policymakers with powerful new tools to make informed intervention decisions."
The authors advocate for rigorous benchmarking of AI models and stress the need for strong collaboration between governments, industries, academia, and society to develop sustainable and effective AI-driven solutions for public health.
