Over the next 2-5 years, AI will transition from being a diagnostic support tool to becoming core public health infrastructure, particularly in disease surveillance, early detection, and delivering healthcare for the masses. In high-burden, resource-constrained settings like India and other lowand middle-income countries (LMICs), AI must function as an embedded, intelligent layer within national health systems rather than a standalone technology.We see this evolution through two complementary tracks. First, AI-powered screening tools like qXR are enabling structured, population-scale early detection of tuberculosis and other lung conditions. qXR can autonomously interpret chest X-rays, prioritise high-risk cases, and standardise reporting across thousands of screening sites. Second, with our AI co-pilot for frontline health workers in LMICs, we are extending AI beyond imaging into primary healthcare delivery. It supports digitised symptom collection, clinical protocol adherence, and real-time decision support. The solution frees up time for patient interaction while simultaneously strengthening data-driven public health planning.The tasks we most want AI to handle are high-volume and repetitive processes including image interpretation, triaging and symptom documentation. When embedded into routine programmes, these capabilities create continuous surveillance systems that flag risk earlier and provide policymakers with actionable intelligence at district and national levels.
.
Scalable systems that serve entire populations
Looking ahead, we want AI systems that are predictive and accountable. The future lies in combining screening intelligence with AI embedded tools. By embedding early detection and intelligent decision support into routine workflows – from rural primary health centres to national disease programmes – AI can help build scalable systems that serve entire populations responsibly.