More and more people are asking OpenAI's ChatGPT and other LLMs questions about their health, and are often finding that chatbots can provide very useful medical insights.
KJ Dhaliwal (pictured left), who sold South Asian dating app Dilmil for $50 million in 2019, said he had been thinking about the inefficiency of the U.S. healthcare system since he was a child, working as a medical interpreter for his parents, and saw the advent of the LLM as an opportunity to do something about it.
In May 2024, he launched Lotus Health AI, a free primary care provider available 24/7 in 50 languages. On Tuesday, Lotus announced it had raised $35 million in a Series A round co-led by CRV and Kleiner Perkins, bringing the total amount raised to $41 million.
People already consult AI about their health, but Lotus goes a step further and goes beyond chat to facilitate actual medical care, including diagnosis, prescriptions, and specialist referrals.
In essence, Lotus is building AI doctors that function just like real medical practices, with licenses to practice in all 50 states, malpractice insurance, HIPAA-compliant systems, and full access to patient records.
The main difference is that much of the work is done by AI, which is trained to ask the same questions as doctors.
Because AI models are prone to hallucinations, the company always has human doctors certified by top medical institutions such as Stanford University, Harvard University, and UCSF review final diagnoses, test orders, and prescriptions.
tech crunch event
Boston, Massachusetts | June 23, 2026
Similar to OpenEvidence, Lotus has developed an AI model that integrates the latest evidence-based research with a patient's medical history and clinical answers to generate a treatment plan.
“The AI is giving advice, but a real doctor is actually approving it,” Dhaliwal told TechCrunch.
Lotus recognizes the limitations of virtual care. In the event of an emergency health issue, Lotus will direct patients to the nearest urgent care center or emergency room. The platform will also refer patients to an in-person doctor if a physical exam is needed, Dhaliwal said.
Outsourcing critical parts of healthcare decision-making to AI is an ambitious gamble given the regulatory hurdles in healthcare. For example, doctors are restricted to seeing patients only in the states in which they are licensed.
“There are a lot of challenges, but it's not about SpaceX sending astronauts to the moon,” said CRV general partner Saar Ghar, who led the deal and joined the company's board of directors.
Gur (pictured right), an early investor in DoorDash, Mercury and Ring, believes the telemedicine framework established during the pandemic and recent advances in AI will allow Lotus to overcome many of the existing regulatory and engineering hurdles.
“That's a big swing,” Garr said. But for investors like Garr, that's the appeal. Lotus is fundamentally rethinking the entire primary care model.
At a time when primary care doctors are in short supply, Lotus claims it can see 10 times more patients than traditional practices, even if each visit is limited to 15 minutes.
This startup isn't the only one developing AI doctors. Lightspeed-backed Doctronic is another competitor. Lotus differentiates itself, at least for now, by offering its entire continuum of care completely free.
Dhaliwal said the eventual business model could include sponsored content and subscriptions, but for now the focus is entirely on product development and patient attraction rather than revenue.