Dr Elaine Hamm
Executive Director, AAIH
Technical progress is no longer the problem. To unlock AI’s full potential, healthcare must fix the systems around it.
Artificial intelligence is not new to healthcare. For decades, researchers have used predictive algorithms to accelerate drug development, from compound screening to modelling disease biology. What is new is the scale, adaptability and promise of today’s AI systems to fundamentally transform how we discover, test and deliver care.
AI innovation blocked by fragmentation
At the Alliance for Artificial Intelligence in Healthcare (AAIH), we have worked closely with stakeholders across industry, academia and government to examine what is working and what still needs fixing. Our recent work on AI in drug development, agentic systems and the absence of a unified patient record all point to the same conclusion: technical progress is outpacing infrastructure.
Our members are optimising trials, identifying safety signals and generating novel therapeutics. However, too often, these tools are deployed in isolation. Real-world data is siloed. Failed trials are hidden. Regulatory expectations are opaque. In our work on unified patient records, we found the same pattern: progress is held back not by capability, but by fragmentation.
Artificial intelligence is
not new to healthcare.
From proof to practice
We are seeing real-world success. Companies are using platform models to reduce drug development timelines from four years to 18 months. Startups are able to identify potential safety signals before they happen. Yet, without access to failure data, safety predictions hit a ceiling. Without consistent incentives, explainable AI remains optional. Without infrastructure, we cannot move from proof of concept to standard of care.
This is not just a technical challenge; it is a policy and governance challenge. If AI is to move from experimental to essential, the global health ecosystem must align principles, validation frameworks and shared expectations.
Healthcare and biotech offer a head start. We already have experience with regulatory oversight, risk management and evidence-based deployment. The next phase of AI adoption will depend on building shared infrastructure — not just locally, but globally. At AAIH, we are committed to helping convene the conversations and collaborations necessary to make that possible.
