Nvidia and Hippocratic AI partner on AI nurses

Nvidia and Hippocratic AI are collaborating on a new project to deploy "AI healthcare agents," which will be powered by Nvidia's powerful H100 Tensor Core GPUs and Hippocratic's healthcare-focused large language model (LLM).

"Voice-based digital agents powered by generative AI can usher in an age of abundance in healthcare, but only if the technology responds to patients as a human would," said Kimberly Powell, Nvidia's VP of Healthcare.

According to Hippocratic's researchers, the company's LLM is trained "on a massive collection of proprietary data including clinical care plans, healthcare regulatory documents, medical manuals, drug databases, and other high-quality medical reasoning documents."

So far, Hippocratic's AI healthcare agents have been tested by more than 1,000 RNs and 100 licensed physicians in the United States. The company is also working with over 40 "beta partners" to test the AI on different responsibilities, including chronic care management, wellness coaching, health risk assessments, and more.

According to an internal survey, Hippocratic claims that its AI nurses perform better than human nurses on bedside manner and education and are close on satisfaction. Other company data also shows that the AI program performs better than human nurses on several tasks, including:

• Identifying a medication's impact on lab values (79% AI nurses vs. 63% human nurses)
• Identifying condition-specific disallowed over-the-counter medications (88% vs. 45%)
• Correctly comparing a lab value to a reference range (96% vs. 93%)
• Detecting toxic dosages of over-the-counter drugs (81% vs. 57%)

Nvidia and Hippocratic have also touted AI agents to ease current labor shortages in the healthcare industry. Currently, the World Health Organization estimates that there will be 10 million fewer health workers by 2030, mostly in low- and middle-income companies.

According to Hippocratic, its AI program is much more cost-effective than real nurses and is "the only scalable way" to reduce the shortfall of healthcare workers. The cost of operating Hippocratic's AI agents is $9 an hour. In comparison, the median hourly pay for RNs in 2022 was $39.05, according to the U.S. Bureau of Labor Statistics.

"With generative AI, we have the opportunity to address some of the most pressing needs of the healthcare industry," said Munjal Shah, cofounder and CEO of Hippocratic AI. "We can help mitigate widespread staffing shortages and increase access to high-quality care – all while improving outcomes for patients. NIVIDIA's technology stack is critical to achieving the conversational speed and fluidity necessary for patients to naturally build an emotional connection with Hippocratic's Generative AI Healthcare Agents." (Zeff, Gizmodo, 3/19; Pugachevsky, Business Insider, 3/22; Paul, Popular Science, 3/19; Dumas, FOXBusiness, 3/21; McClure, New Atlas, 3/25)

AI can't replace RNs — but here's how it can help

By Coltin Ball and Monica Westhead

As health systems continue to seek sustainable solutions to ease staffing shortages and bridge a widening experience-complexity gap, an AI-based solution aimed at replacing nurses was inevitable.

In a broader sense, Nvida and Hippocratic AI's partnership marks notable progress in emphasizing innovation in nursing, which has historically received underinvestment relative to other healthcare professionals. However, it should go without saying: AI cannot comprehensively fulfill a nurse's full scope of practice. 

AI shouldn't replace RNs

When considering what tasks AI may offload from a nursing care team, it is important to account for the increasingly complex nature of nursing care, as well as AI's limitations. For example, based on the data it was trained on, AI can be biased, and can make recommendations based on incomplete, inaccurate, or outdated practices. 

An AI "nurse" may struggle to adapt to individual patient needs, preferences, and unique colloquialisms, potentially increasing instances of missed care, or leading to oversight of deterioration — which can be financially, socially, and ethically costly for a health system. While AI can aid in certain tasks, it cannot offload the complete scope of nursing practice.

Nursing is much more than completing tasks — it involves building relationships and critical thinking.

Why health systems should still invest in AI and nursing technology

While AI can't replace RNs, it certainly has the potential to alleviate below-license work, streamline workflows, and enhance the potential for quality patient outcomes. AI applications like automated data entry of vital signs, clinical decision-making support, and AI-assisted triaging can reduce and support a nurse's workload.

Health systems should not underestimate the potential of AI to revolutionize nursing care delivery through seemingly ever-evolving, never-ending use cases. To avoid falling behind competitors, organizations must carefully strike a balance between being at the forefront of innovation with an incremental, and critically-evaluated strategy.

As part of ongoing efforts to address the nursing crisis, Advisory Board is conducting research on innovative solutions in nursing. Interested in adding to the conversation? Our team would love to hear your thoughts as we continue our research. Please email Coltin Ball at ballc@advisory.com or Marissa Goodall at goodallm@advisory.com to schedule a research interview.

Leaders are searching for new solutions to support clinicians and fill staffing gaps at their organizations in the face of clinician burnout and labor shortages. This decision guide serves as a resource for leaders to quickly index the state of workforce technology and how it can benefit their organizations.