Advancing healthcare and scientific discovery with AI

Last week, at the Lake Nona Impact Forum for advancing global health, I discussed the potential of AI to meaningfully improve healthcare and advance science. Our recent AI breakthroughs provide unprecedented opportunities to make healthcare more accessible, personalized and effective for everyone, and to significantly accelerate scientific discovery. Here’s an update on our progress, how we’re collaborating with partners to bring AI to global healthcare settings and our recently announced AI co-scientist.

AI is making accurate health information more accessible

Google is often the first place people turn to when they are looking for answers to health-related questions, so we strive to make sure everyone has access to relevant, high-quality health information in their moment of need. Using Lens, people can take a picture to search skin conditions that are visually similar to what they see on their own skin, and on YouTube, we piloted AI tools with health creators and organizations like Cleveland Clinic to make it easier for them to publish authoritative, high-quality content.

For care providers, we launched MedLM and Search for Healthcare which can provide answers to medical questions. These are available on the Google Cloud Vertex AI platform, helping clinicians make more informed decisions and helping patients receive the precise care they need. Our research into medical factuality ensures that health-related content generated by language models is as reliable and grounded in factual sources as possible.

Generative AI is paving the way for personalized healthcare

With advancements in multimodality and conversational AI, we’re able to reimagine patient care and how it could be personalized for everyone, with an emphasis on preventive healthcare.

From X-rays to digital health records, medicine is multimodal. Building on our MedLM research, we developed Med-Gemini, next-generation models for healthcare which have Gemini’s superior multimodal and reasoning capabilities and are fine-tuned on de-identified medical data. In published research, Med-Gemini achieved 91.1% accuracy on U.S. medical exam-style questions, and we demonstrated how the models can effectively interpret 3D scans or answer complex clinical questions.

We’re researching how AI systems could serve as conversational diagnostic partners in clinical settings, using Articulate Medical Intelligence Explorer (AMIE), a research AI system optimized for diagnostic reasoning and conversations. It is designed to take a “clinical history” and ask intelligent questions to help derive a differential diagnosis and handle discussions with empathy, including in subspecialist domains.

Mobile and wearable devices are another promising area where generative AI models could provide personalized insights for both healthcare and wellness, using data such as step count and heart rate. We designed the Personal Health Large Language Model, another fine-tuned version of Gemini, which can interpret sensor data and generate insights and recommendations about an individual’s sleep and fitness patterns.

AI is improving health outcomes globally

Early diagnosis of disease is critical for improving health outcomes. Over the past decade, we’ve tapped into AI’s imaging and diagnostic capabilities and developed AI models to help detect diseases including breast cancer, lung cancer and diabetic retinopathy. Through partnerships, we’re now bringing these solutions at scale to clinical settings so that more patients can benefit from timely and accurate screenings. The impact is particularly profound in low-resource medical settings and countries with fewer specialist doctors per capita. Over the next decade, our health-tech partners in India and Thailand aim to deliver 6 million diabetic retinopathy screenings at no cost to patients, and Apollo Radiology International will build on our AI models to provide 3 million free screenings across India for tuberculosis, lung cancer and breast cancer. Adding to our various initiatives to address maternal health in Africa, we’re also developing an ML model for cardiotocography, used to predict fetal well-being, and exploring its utility in medical settings with limited resources.

We’re also laying the technological foundations for broader access to healthcare. Our Health AI Developer Foundations include open-weight models and resources to help developers build AI models for healthcare more efficiently. And solutions powered by our Open Health Stack (OHS) — a suite of open-source tools that makes it easier for developers to create next-gen digital health solutions for healthcare workers — have already been deployed across regions in Africa, South Asia and Southeast Asia to support frontline healthcare workers serving millions of patients.

AI is accelerating scientific discovery

Medicine is rooted in science. Building on AI’s ability to synthesize information and perform complex reasoning tasks, we’re exploring how it could augment scientific and biomedical discovery through our work on AI co-scientist, a multi-agent AI system based on Gemini 2.0. AI co-scientist is designed to function as a collaborative tool for scientists. It’s intended to uncover new, original knowledge and help scientists formulate novel research hypotheses and proposals, building upon prior evidence and tailored to specific research objectives. It has already demonstrated potential in areas such as drug repurposing for acute myeloid leukemia, proposing hypotheses for novel treatment targets for liver fibrosis and explaining mechanisms of horizontal gene transfer underlying antimicrobial resistance — each of which is a complex application and presents a different set of challenges.

We continue to realize the incredible potential of AI to advance science and improve, personalize and democratize access to healthcare. The “magic cycle” where we achieve research breakthroughs and translate them to real world impact is accelerating and expanding in scope. We’ll pursue this opportunity responsibly, in collaboration with global partners, and continue to share our research — in 2024, we published more than 50 papers in which we shared cutting-edge health research, and we recently shared our 2025 Health Impact Report. Ultimately, we believe AI will continue to help advance healthcare and science for the benefit of billions of people.