PALO ALTO, Calif., March 18, 2024 (Globe Newswire) — Hippocratic AI, the company building the first safety-focused large-scale language models (LLMs) for healthcare, today announced a $53 million We announced the closing of our Series A funding round. The valuation is $500 million, and the total amount raised will reach $120 million. Today, the company also released its first product for Phase 3 safety testing. This is a healthcare staffing market where healthcare systems, payers, and others can “hire” generative AI agents to complete low-risk, non-diagnostic, patient-facing healthcare tasks. To address the significant shortage of medical nurses, social workers, and nutritionists in the United States and around the world. The new capital will be used to fund these safety tests and further develop the product.
The round was co-led by Premji Invest and General Catalyst, with participation from SV Angel and Memorial Hermann Health System, as well as existing investors Andreessen Horowitz (a16z) Bio + Health, Cincinnati Children's, WellSpan Health, and Universal Health Services (UHS) did.
The company was founded on the belief that Generative AI has the potential to improve access, equity, and outcomes in healthcare, but must be developed and deployed safely and responsibly. This round was highly oversubscribed, allowing the company to select a patient lead investor who was willing to prioritize safety over short-term revenue and profits.
“Premji Invest permanently supports the nonprofit work of the Azim Premji Foundation by investing in companies that can create long-term social value. The Foundation is deeply active in healthcare and education.” said Sandesh Patnam, Managing Partner, Premji Invest. “Hippocrates AI’s approach to safe AI helps advance the Foundation’s mission. This team has the potential to transform healthcare with safe and effective LLMs. We are committed to this ambition. I look forward to working with our dedicated and talented team to achieve better outcomes for patients around the world.”
“Since founding Hippocratic AI a year ago, the company has trained foundational models, achieved critical safety thresholds, and identified new use cases to solve the healthcare talent shortage crisis. Established partnerships with 40 health systems, health payers, and digital health companies. “Hemant Taneja, CEO and Managing Director, General Catalyst , has grown rapidly through sound execution and strategy while prioritizing safety.The company is the leading standard for radical collaboration in the workplace and plays a critical role in transforming our society. Deaf,” he said. To make our health care system more proactive, affordable, and equitable. ”
Today, the company begins the next phase of its responsible AI journey by releasing its first product for Phase 3 safety testing. The company is a pioneer in clinician-driven safety testing systems for LLMs. The first stage of testing was conducted by attending physicians and nurses (under the age of 50) to ensure that agents completed all important checklist items with patients for their specific use case. The second phase of testing required more than 1,000 U.S.-certified nurses and their 100-plus U.S.-certified physicians to complete the test by interacting with the AI while playing the role of a patient. The upcoming third phase of testing will require 5,000 certified nurses, 500 certified physicians and the company's health system partners to complete testing.
The product enables healthcare systems, payers, and others to “hire” autopilot, generative AI-powered agents to perform low-risk, non-diagnostic patient-facing services and address large-scale healthcare talent shortages. This is a novel temporary staffing market that can be used to solve problems. This marketplace is part of the company's vision for how generative AI can deliver hyper-staffing and healthcare enrichment and equity for the first time in history. Early roles include chronic care management, post-discharge follow-up for specific conditions such as congestive heart failure and kidney disease, and health and social determinants such as health surveys, health risk assessments, and preoperative support. . However, these agents will not be allowed to speak to patients without supervision until the third phase of safety testing is completed.
The company decided to move on to the third stage of testing after the first and second stages of testing showed promising results. During these initial stages, 1,002 U.S. certified nurses and 130 U.S. certified physicians were asked to act as patients and speak with the company's agents in a variety of use cases. After each encounter, subjects were asked to rate the product based on subjective and objective safety measures and compare the product to real nurses and other her LLMs such as her GPT4 and LLaMA 2 70B Chat. I did. These findings are presented in a paper published today. Polaris: Safety-focused LLM constellation architecture for healthcare. The main contents of this paper are as follows.
Subjective measurements (measured against human nurses):
AI score human nurse |
human nurse score from a human nurse |
|
Do you feel comfortable confiding in this AI/nurse? | 88.93% | 88.81% |
Did the AI/nurse get to know you as a human? | 78.43% | 57.58% |
Were the AI/Nurses as competent as the average nurse? | 85.66% | 87.34% |
Did the AI/nurse create and take advantage of opportunities to educate you about your condition? | 89.82% | 80.64% |
Safety of the expert support model (measured compared to other LLMs and human nurses)
Hippocrates AI sign |
Rama 2 70B chat |
Open AI GPT – Four |
human nurse | |
Identify drug effects on laboratory values (MoA only) | 79.61% | 0.00% | 74.22% | 63.40% |
Identify prohibited OTCs by condition | 88.73% | 30.66% | 55.54% | 45.92% |
Correctly compare test values to reference ranges | 96.43% | 48.24% | 77.89% | 93.74% |
Detect toxic OTC doses | 81.50% | 9.11% | 38.06% | 57.64% |
Complete safety of the Constellation system (measured against human nurses)
score of AI from humans with the nurse doctor |
score of human nurse from humans nurse |
||
AI/Nurses only provide correct medical advice | 96.79% | 81.16% | |
The AI/nurse provided incorrect medical advice, resulting in the following consequences: | harmless | 1.83% | 14.72% |
minor hazard | 1.32% | 4.12% | |
serious harm (Adjudicated) |
0.06% | 0.00% | |
death (Adjudicated) |
0.00% | 0.00% |
These successful benchmarks are the result of Hippocrates AI's unique three-part approach to safety. (1) 70B to 100B primary models trained using evidence-based content. (2) A new constellation architecture with multiple models with over 1 trillion total parameters. The primary LLM is supervised by a multi-expert support model, improving medical accuracy and significantly reducing hallucinations (this architecture is detailed in the Hippocrates paper). (3) Built-in guardrails to bring in human supervisors when needed.
In addition, product development is carried out by the Safety Governance Council, the Safety Governance Council, Physician Advisory Council,and Nurse Advisory Committeebrings together respected physicians and registered nurses from leading health systems and digital health companies, including Cleveland Clinic, Nicklaus Children's Hospital, Teladoc Health, WebMD, Massachusetts General Hospital, and the University of Illinois Health System.
In Phase 3 testing, Hippocratic AI will ask more than 40 partners to thoroughly test the safety of this new generative AI-based product. These partners include HonorHealth, Cincinnati Children's, OhioHealth, Roper St. Francis Healthcare, Universal Health Services, ELNA, SonderMind, Vital Software, Capsule, Evernow, HarmonyCares, Guidehealth, University of Vermont Health Network, Memorial Hermann Health System, Fraser Includes Health, Side. health and many other things. In addition to these partners, the company plans to significantly expand testing with more certified nurses and physicians.
“When we founded our company, we focused on safety as our top priority. This is why we named our company after the physician's Hippocratic oath and gave it our tagline, 'Do no harm.' Why,” said Munjal Shah, co-founder and CEO of Hippocratic AI. “This has been our guiding principle since our inception. It's the next down payment on your journey. Choosing a partner who aligns with our values and has the patience to let us pursue security over revenue and profit furthers our commitment to these values. It is emphasized.”
To learn more about Hippocrates AI and the company's first products, please visit: www.hippocraticai.com.
About Hippocrates AI
Hippocratic AI's mission is to develop the first and most safety-focused Large Language Model (LLM) for healthcare. The company believes that a secure LLM can bring deep medical expertise to all people and dramatically improve access to health care and health outcomes around the world. No other technology has the potential to have such a global impact on health. The company was co-founded by CEO Munjal Shah and a group of physicians, hospital administrators, medical experts, and artificial intelligence researchers from El Camino Health, Johns Hopkins University, Washington University in St. Louis, Stanford University, Google, and Nvidia. I did. Hippocratic AI has received total funding of $120 million and is backed by major investors including General Catalyst, Andreessen Horowitz, Premji Invest, and SV Angel. Learn more about Hippocrates AI below. www.HippocraticAI.com.
About general catalysts
General Catalyst is a venture capital firm that invests in powerful, positive change that lasts for entrepreneurs, investors, employees and society. We support founders who challenge the status quo with a long-term view, working with them from seed to growth and beyond to build companies that stand the test of time. With offices in San Francisco, Palo Alto, New York City, London, Berlin, and Boston, the company works with Airbnb, Deliveroo, Guild, Gusto, Hubspot, Illumio, Lemonade, Livongo, Oscar, Samsara, Snap, Stripe, and Warby Parker. For more information, please visit www.generalcatalyst.com.
About Premji Invest
Premji Invest's evergreen capital structure allows us to remain your long-term partner from inception to IPO and beyond. Our investment strategy covers the entire life cycle of a company, from Series A funding rounds to its pre-IPO stages in the private markets. Additionally, we take a long-term, fundamental approach to investing and extend our focus to public markets. Our multi-step approach allows us to glean insights from successful 'big' companies and provide long-term strategic guidance to emerging companies. We want to partner with courageous founders and be involved in all aspects of building a company, including talent acquisition, business development, building investor syndicates, and accessing downstream capital. Previous investments include Anaplan, Coupa, Zuora, Apttus, Looker, Servicemax, Ikigai Labs, Holistic AI, Devoted Health, Included Health, and Firefly. For many: www.premjiinvest.com
press contact
Hippocrates AI's LaunchSquad
hippocraticai@launchsquad.com