Hippocratic AI, a company building a safety-first focused artificial intelligence large language model for the healthcare industry, announced Monday it has raised $53 million in new funding increasing the company’s valuation to $500 million.

The Series A funding round was co-led by Premji Invest and General Catalyst with participation from SV Angel and Memorial Hermann Health System. Existing investors Andreessen Horowitz, WellSpan Health, and Universal Health Services also joined the round. The funding brings the total raised by the company to $120 including $50 million raised by the company in a seed round in May 2023.

Founded by a team of medical professionals, hospital administrators, physicians and medical researchers, Hippocratic is named after the oath taken by doctors to “do no harm.” The artificial intelligence model the company developed is designed to be deployed safely and responsibly in healthcare use cases for the benefit of hospital administrators and clinicians to aid in patient care outcomes and streamline support.

With the announcement, Hippocratic AI released its first product for “phase three safety testing.” It is a staffing marketplace for healthcare where health systems, payors and others can discover generative AI agents, that can complete low-risk, non-diagnostic, patient-facing healthcare tasks.

The company said that the objective of these generative AI agents and chatbots would be to help solve the massive shortage of nurses, social workers and nutritionists that the healthcare industry is currently facing in the U.S. and worldwide. Examples include a virtual agent handling questions regarding chronic care management, medication checks and post-discharge follow-up for specific conditions such as kidney failure and congestive heart failure. However, the AI agents cannot speak to patients unsupervised until phase three of the safety testing is completed.

Phase one testing was conducted by just under 50 physicians and nurses to make sure that the AI agents complied with best practices and standards when addressing patient care. Phase two testing required a large pool of test participants including more than 1,000 licensed U.S. nurses, and more than 100 U.S. doctors talking to the AI while acting as a patient. Within phase three, those numbers rose above 5,000 nurses and 500 doctors.

“When we started the company, we prioritized safety as our top value,” said Munjal Shah, co-founder and chief executive of Hippocratic AI. “This has been our guiding principle since the company’s founding. Our focus on safety testing our product in multiple phases and transparent publication of the results for everyone to see is the next down payment in this safety-first journey.”

According to Hippocratic AI, the company’s LLM outperformed other LLMs such as OpenAI’s GPT4 and Meta Platform Inc.’s Lama 2 70B Chat on subjective and objective safety measures as compared by the nurses and doctors in outgoing survey assessments after each safety phase. The company said that the safety capability is due in part to the primary LLM being supported and supervised by “multiple specialist support models” that improve medical accuracy and “substantially reduce hallucinations.”

Additionally, product development is guided by several oversight groups aligned with guiding the safe design and development of the model. These groups include a governance council, a Physician Advisory Council and a Nurse Advisory Council. The three oversight groups bring together registered physicians and nurses from various health systems including Cleveland Clinic, Nicklaus Children’s Hospital, Teladoc Health, WebMD, Massachusetts General Hospital and the University of Illinois Health System.

Photo: Pixabay