Imane Motaib, professor of endocrinology, diabetology and metabolic diseases at the University Mohammed VI Polytechnic, examines the evolving role of AI in reshaping diabetes care.
Diabetes is one of the defining health challenges of our time. In the past three decades, the number of people living with the disease has quadrupled worldwide. By 2050, Africa alone is projected to see a 142% increase in prevalence, with 60 million adults affected. This trajectory is unsustainable. It places enormous strain on healthcare systems, threatens economic productivity, and affects the lives of millions of families.
What is needed is a shift towards earlier detection, personalised care, and scalable innovation. Artificial intelligence (AI) is uniquely positioned to deliver this shift, providing new tools to predict, prevent, and treat diabetes more effectively.
One of the most powerful applications of AI is the ability to detect early warning signs of diabetes. By analysing large and complex datasets, from electronic health records and lab results to imaging scans, AI systems can identify patterns invisible to the human eye.
For example, AI models trained on longitudinal health data can flag subtle changes in blood glucose trends, BMI, and metabolic markers. These predictive tools support proactive interventions, such as lifestyle coaching or targeted therapies, which can delay or prevent disease onset.
AI is also being used to forecast complications like cardiovascular disease, kidney damage, and diabetic retinopathy. By anticipating these risks earlier, clinicians can tailor preventive strategies and reduce long-term healthcare costs.
Towards personalised, real-time care
For those living with diabetes, management has relied on periodic clinical visits and self-reporting, approaches that provide only a partial picture of disease progression. AI is changing this by enabling continuous, personalised monitoring.
Wearable devices and continuous glucose monitors (CGMs) collect real-time data on blood sugar levels, physical activity, and dietary habits. AI algorithms analyse this data instantly, offering tailored recommendations and alerts. For example, some systems predict glucose spikes and suggest insulin adjustments or dietary changes before symptoms occur.
Importantly, this empowers patients. With timely feedback and personalised insights, individuals can take greater ownership of their health, while clinicians benefit from a richer and more accurate understanding of disease management outside the clinical setting.
AI’s impact is not limited to frontline care. It is revolutionising the pace and scope of research. Drug discovery has been a slow and resource-intensive process, taking years to move from concept to trial. AI shortens timelines dramatically by analysing vast molecular datasets, identifying potential compounds, and predicting their effectiveness.
In the context of diabetes, this means accelerating the development of new therapies, from glucose-lowering agents to treatments that address complications. Equally, AI can support efficient clinical trials by identifying suitable participants and monitoring outcomes with greater precision.
The result is a faster, more targeted pipeline of innovation, critical in the ongoing search for therapies that improve outcomes and, ultimately, work towards a cure.
A major advancement in diabetes technology is the development of AI-powered closed-loop insulin delivery systems, often referred to as artificial pancreas devices. These systems integrate CGMs with insulin pumps and AI algorithms to automatically adjust insulin delivery in real time.
By continuously analysing glucose levels and predicting future trends, closed-loop pumps help maintain optimal glycaemic control with minimal manual input. This not only improves clinical outcomes but also enhances quality of life and significantly reduces anxiety for patients, who no longer need to make frequent dosing decisions.
AI is also playing a growing role in therapeutic education, particularly for patients with type 2 diabetes. AI-powered chatbots offer interactive, cost-effective, and accessible guidance on disease management, lifestyle changes, medication adherence, and more.
Virtual assistants answer questions, provide reminders, and deliver personalised educational content in multiple languages and formats. For patients in remote or underserved areas, chatbots offer a scalable solution to bridge gaps in diabetes education and support.
Building global solutions
Diabetes is a global challenge that requires a global response. AI creates opportunities for collaboration that transcend borders. Shared data platforms and research networks can accelerate discovery and ensure that breakthroughs benefit diverse populations rather than being concentrated in a handful of regions.
Equitable access is, however, essential. The steepest increases in diabetes are expected in low- and middle-income countries, where healthcare resources are often limited. AI can help bridge these gaps if its deployment is guided by principles of fairness, accessibility, and cultural relevance.
The trajectory of diabetes prevalence is deeply concerning, but not inevitable. AI can fundamentally change the way we approach the disease: from predictive analytics that enable earlier intervention, to personalised care that adapts in real time, to accelerated discovery that expands treatment options.
These capabilities do not replace clinical expertise; they enhance it. By combining the power of AI with the knowledge of healthcare professionals, we can transform diabetes care from reactive crisis management to proactive, patient-centred prevention and treatment.
As healthcare systems continue to grapple with rising demand and constrained resources, the integration of AI is a necessity. The challenge is to ensure that its promise is realised equitably and sustainably, so that every individual living with or at risk of diabetes can benefit.
