How Can Trump’s Announced $500 Billion Investment In AI Change Healthcare?

President’s Trump recent announcement of a $500 billion investment in AI infrastructure will change the game for multiomics, a very complex, but very promising, medical research field. Despite buzzwords of personalized care and precision medicine, we, physicians, are still delivering healthcare imprecisely, practicing largely by old perceptions of broad categorization. Now, science is finally catching up with the vision of personalized care, largely thanks to multiomics developments.

Multiomics research refers to an integration of different medical data layers, such as genomics, proteomics, metabolomics and microbiomics, as each of these ‘omic’ layers represents the next runner in the relay race of the human body, from the policy level (the DNA) to on-the-ground implementation with protein production, for example, and up to our body’s reaction to our diet, with metabolomics and microbiomics.

Integrating and interpreting such massive and diverse quantities of data at scale requires AI, and the latter needs high-end infrastructure, such as data centers and dedicated computer chips. This is the goal of the newly announced Stargate initiative, led by a collaboration of giants including OpenAI, SoftBank and Oracle. The news about the new AI company, not specifically targeting health research but rather the infrastructure that will support it, is preceded by another collaboration announcement just a week before between Illumina, a key player in genomic sequencing, and NVIDIA, to allow multiomic analysis at scale, for research and drug discovery. And this is just the start. The global multiomics market is expected to flourish in the next five years, from $2.7 billion in 2025 to $5.1billion by 2029.

The next uprising research field is polygenic risk score. Although PRS has yet to be implemented in clinical practice, it shows remarkable potential. The eMERGE scientific network, for example, a group of studies across the US funded by the NIH, has published results of PRS that evaluate one’s risk for conditions such as asthma, atrial fibrillation, chronic kidney disease, diabetes, and coronary heart disease.

How will knowing our genetic risk impact our lives?

Our genome contains about 3 million DNA base pairs of a complex coding system, overwhelmingly similar between all humans. The small differences between people, or genomic variants, are what sets each person apart; some of them are decoded to physical characteristics and personality traits, while others increase or decrease the risk for developing diseases.

Some medical conditions are single-gene diseases, but many are more complex, influenced by a combination of genetic variants (or poly-genic), of hundreds or even thousands of variants, with environmental and lifestyle influences, such as air pollution, diet, stress, or smoking. Recent developments in genetics and technological and computational advancements have led scientists to take a comparative approach, comparing a group of people with a disease to another group without inferring the significant combination of genetic changes that might lead to that disease. Each genetic variant might have a small impact on its own, but together, they form a pattern that increases an individual’s likelihood of developing a condition.

PRS, a statistical score, is a number that reflects one’s genetic predisposition for medical conditions. This field is emerging now given that humanity has already created a large enough reservoir of sequenced genetic data, computation power to process it, and advanced statistical models to understand it. Combining PRS with clinical data has an advantage in risk assessment. A new study published in JAMA leveraged genetic data to better identify people with undiagnosed COPD, a chronic lung disease associated with smoking and pollution, constituting the fourth leading cause of death worldwide. Researchers showed that integrating PRS with clinical questionnaires led to higher detectability rates than questionnaires alone.

Risk assessment, not prophesy

PRS is all about probabilities, not absolute risks. This is why we should use PRS as another layer of our multiomics analysis, another piece of our healthcare journey. Also, many of the existing PRS lack diversity, as they were developed on the genetics of people of European descent. This means they need to be validated and adapted to people of different ancestry to allow universal adoptability.

The future of multiomics in healthcare

Leading the next leap in AI-driven healthcare requires brain power and a huge investment of money. It also needs the courage to implement responsibly. In the near future, integrating multiomics into clinical practice will give deeper insights into patients’ health trajectories and what influences them, helping physicians and empowering patients to navigate their health in the right direction with earlier interventions, tailored treatments, and more effective prevention strategies. This is how medicine will finally shift from reactive to proactive care.