This article is part of our Future of AI series from Imagination in Action 2025 Silicon Valley Summit — where founders, leaders, and investors explored what’s next for AI. Explore the magazine.
Artificial intelligence, long imagined as a potential tool for diagnostics or administrative efficiency, is now reshaping the very architecture of health care. From personalized mental health support to predictive ultrasounds and drug discovery, AI is not just augmenting medicine but transforming it.
With the growth of AI, health care may start to buck its reputation as a slow adopter of new technologies. The use cases are rapidly spreading, with models capable of interpreting medical images and CT scans. For example, AI has already been used to successfully diagnose diabetic retinopathy from retinal images with greater sensitivity than human ophthalmologists.
The problems AI can solve are no longer theoretical. Some of these AI apps are already being used in hospitals and clinics. As AI models’ speed and precision increases, health care stakeholders are no longer looking at merely incremental gains and process enhancements, but paradigm shifts in how they practice medicine.
Developing trust with providers
The winners in health care transformation won’t be those with the most code or capital. They’ll be the ones who earn trust.
When Robert Bunn, founder of Ultrasound AI, wanted to transition from the oil industry to women’s health care, his challenge was to get doctors to trust an AI tool that could predict what they couldn’t: the actual delivery date of a baby. Ultrasound AI uses ultrasound images of the fetus as early as six weeks into the pregnancy and can deliver remarkably accurate results. This enables providers to know if a premature birth or miscarriage is coming, and intervene earlier than ever.
“I got into this for very personal reasons,” Bunn said. “My wife had numerous miscarriages, and I decided I wanted to apply my skills to solving that problem.”
Doctors initially told him it couldn’t be done—especially not by someone without a medical degree. But after a year of work and clinical validation, his tool has now been in use in multiple countries across South America for the past year.
Using enterprise AI, from automation to doctors’ chatbots
Mudit Garg sees a similar potential to improve automation across health care. He’s the CEO of Qventus, a startup building AI tools for more than 200 hospitals across the country. It is already being used for prepping patients for surgery, scheduling, and releasing patients after care.
For patients, accessing health care can often feel like “a rock being pushed up the hill,” Garg said. Tackling the data problem helps them manage the patient journey with greater ease.
In health systems, the degree of unstructured data and interactions (everything from visits, claims, faxes) far outpaces anything we capture in a structured standpoint, Garg explained. Structured data is organized in a predefined format, making it easy to process and analyze; unstructured data has typically been much harder to understand. Additionally, a host of logistical inefficiencies—missed appointments, redundant paperwork, fragmented records—have plagued health systems.
For patients, accessing health care can often feel like “a rock being pushed up the hill,” Garg said. Tackling the data problem helps them manage the patient journey with greater ease.
In health systems, the degree of unstructured data and interactions (everything from visits, claims, faxes) far outpaces anything we capture in a structured standpoint, Garg explained. Structured data is organized in a predefined format, making it easy to process and analyze; unstructured data has typically been much harder to understand. Additionally, a host of logistical inefficiencies—missed appointments, redundant paperwork, fragmented records—have plagued health systems.
AI agents can now streamline these processes, parsing unstructured data like faxes and handwritten notes, and generate actionable insights. What once required a team of nurses and administrators can be handled by intelligent systems in seconds, for instance to manage surveys and intake forms. It’s also possible to use AI to give providers a more complete picture of patients, Garg said.
“For patients, accessing health care can often feel like a rock being pushed up the hill.”
Mudit Garg, Qventus
Garg sees greater adoption and hunger for these kinds of innovations, especially given the rising costs of health care for both systems and patients. Manually doing everything isn’t sustainable, so these hospitals have been rapidly adopting the platform and its AI features.
At Stanford Health Care, a new AI agent for clinicians helps expedite daily tasks, including chart reviews and asking specific questions regarding patients. ChatEHR, artificial intelligence software developed at Stanford Medicine, allows the staff to interact with patients and data in plain language, much like in ChatGPT. For example, they can ask: “Has this patient ever received therapy?” or “How many times has this patient received a blood thinner?”
As ChatGPT became popular, it became obvious to the Stanford team that it would make sense to create an app for direct care, said Topher Sharp, clinical professor in medicine, care and population health at Stanford Health Care. “For physicians, this would be an ability to augment their search,” Sharp said.
Sharp knew that faculty at Stanford would use ChatGPT if they didn’t have a better alternative. So ChatEHR draws from a variety of data sources, including Epic Systems, a widely-used Electronic Health Record (EHR) software platform, to access patient data. But it does so in a secure environment, ensuring the data isn’t being used to train the model.
“Our core business isn’t AI, but we do want to use it.”
Nigam Shah
“Our core business isn’t AI, but we do want to use it,” added Nigam Shah, professor of medicine at Stanford and chief data scientist at Stanford Health Care.
As with many AI integrations, the end goal is to reduce the cost of health care by reducing and streamlining repetitive tasks. But the broader vision extends far beyond cost savings to improving—and even transforming—the way patients receive care.
Founder Takeaways
Explore The Future of AI | This article is part of our Future of AI series from Imagination in Action 2025 Silicon Valley Summit — where founders, leaders, and investors explored the next revolution of AI. We explored how AI is changing scientific research, creating new startup economics, straining power grids, and challenging us to rethink everything from enterprise software to regulatory frameworks. Dive into the Future of AI magazine to see the full picture.
