What happened
Dr. Munjal Shroff, a psychiatrist in Atlanta, began searching for a healthcare‑specific AI assistant when his wife experienced unexplained symptoms in late 2025. He signed up for Healz AI, a chatbot designed to help cancer patients track diagnoses, treatments, and recovery. The app recommended a second opinion at the MD Anderson Cancer Center, which resulted in a diagnosis of rare T‑cell lymphoma. Throughout her treatment, Shroff used Healz to explore chemotherapy options and to understand medical information, noting that the AI’s suggestions often aligned with his doctors’ recommendations. After months of heavy use, Shroff invested in Healz, joining founders Igor Demishev and Alex Plat as an investor. The company, backed by Recursive Superintelligence co‑founder Alexey Dosovitskiy, reported that its annual recurring revenue reached $100,000 in early September, with a 70% month‑to‑month user retention rate and a user base that is 97% cancer patients.
In the fall of 2025, Dr. Shroff’s wife experienced rashes, itching, fatigue, and swollen lymph nodes. After multiple consultations and inconclusive tests, Shroff sought a digital aid and discovered Healz AI, a chatbot launched in beta the previous year. The app’s recommendation to seek a second opinion at the University of Texas MD Anderson Cancer Center led to a definitive diagnosis of T‑cell lymphoma.
Healz’s functionality includes symptom tracking, document ingestion (one user uploaded 1,500 medical files), and a "root‑cause mode" that asks up to 20 follow‑up questions before offering suggestions. The platform also helps patients locate oncologists, interpret scans, and navigate treatment plans. Shroff used the tool to compare chemotherapy options, and his doctor confirmed the AI’s recommendation as appropriate.
The startup’s founders, Igor Demishev and Alex Plat, emphasized that Healz does not replace physicians but serves as a supplemental resource. They reported that recurring revenue doubled each month from June to August, reaching $100,000 in annual recurring revenue by early September, with a 70% retention rate and daily or weekly engagement from most users.
Following his personal experience, Shroff invested in Healz, joining a funding round that includes Alexey Dosovitskiy, co‑founder of Recursive Superintelligence, a company valued at $4.65 billion. The founders disclosed plans for further fundraising, though pricing and broader access details remain undisclosed.
Source details: businessinsider.com ↗
Why it matters
Healz AI illustrates how specialized AI tools can augment patient advocacy and potentially accelerate critical medical decisions, especially for rare cancers where diagnostic pathways are complex. By prompting a timely second opinion, the chatbot may have reduced diagnostic delay, a factor known to affect outcomes in oncology. The startup’s early revenue growth and high retention suggest a market appetite for AI‑driven patient support, signaling a shift toward AI‑enabled self‑service in healthcare. Moreover, the involvement of a high‑profile investor like Alexey Dosovitskiy underscores growing confidence in AI applications beyond generic large‑language models, highlighting the commercial viability of domain‑specific AI solutions. However, the story also underscores the necessity of human oversight, as the founders and Shroff acknowledge the AI’s occasional inaccuracies and stress that it should supplement, not replace, clinicians.
The case demonstrates a concrete instance where an AI tool directly influenced a critical medical decision, potentially shortening the time to accurate diagnosis for a rare cancer—a scenario where delays can be life‑threatening.
Healz’s early revenue and high retention suggest that patients are willing to pay for AI‑driven health navigation, indicating a viable business model for specialized AI applications in medicine.
The involvement of a prominent AI investor signals confidence in the commercial potential of domain‑specific AI, which may encourage further venture capital interest in health‑focused AI startups.
The founders’ acknowledgment of the AI’s limitations reinforces the importance of maintaining clinician oversight, a key consideration for regulators and healthcare providers evaluating AI integration.
Interactive Mechanism: How It Actually Works
Explore the underlying technology behind this development interactively.
Impossibility results in algorithmic fairness (e.g. Kleinberg et al., Chouldechova) show what?
What to watch next
Healz plans additional fundraising in the coming months, which could expand its user base and product features. Observers should monitor whether the startup secures broader clinical partnerships, integrates with electronic health‑record systems, or obtains regulatory clearances that would enable wider adoption. Tracking the performance of its “root‑cause mode” and memory system in real‑world settings will be key to assessing the tool’s reliability and impact on patient outcomes. Finally, the broader healthcare industry’s response—particularly from hospitals and insurers—will indicate how quickly AI chatbots can move from niche patient‑support tools to mainstream components of cancer care pathways.
Future fundraising rounds could provide the capital needed for product scaling, regulatory compliance, and integration with hospital systems.
Potential partnerships with major cancer centers or health insurers would broaden Healz’s reach and could set industry standards for AI‑assisted patient support.
Regulatory scrutiny, especially regarding AI recommendations in oncology, may affect the startup’s ability to market its tool beyond a beta phase.
User outcomes data—such as diagnostic timelines and treatment adherence—will be essential to validate the tool’s clinical impact and to attract broader adoption.