AI doctors: which startup is ahead?

Which AI doctor startups actually belong in this race?

The serious AI doctor field currently contains ten startups, although only four are making a strong claim to lead the category today.

We include companies whose software influences a real clinical step: assessing symptoms, suggesting a diagnosis, shaping treatment, managing medication, supporting a physician’s decision or delivering care through a licensed medical practice.

K Health, Counsel Health and Lotus Health AI use artificial intelligence inside physician-supervised consultations. Doctronic has gone further in one narrow area, with an AI system authorized to renew certain prescriptions in Utah without a doctor reviewing each case. UpDoc manages insulin between appointments, although a clinician must first set the patient’s treatment limits.

Ada Health, Ubie and Infermedica mainly assess symptoms and direct people toward the right care. They can affect medical decisions, but they usually stop before diagnosis or treatment. OpenEvidence helps doctors answer clinical questions but does not treat patients itself. Hippocratic AI speaks directly with patients at huge scale while deliberately avoiding diagnosis and prescribing.

We exclude Abridge, Nabla and Suki because they mostly document medical visits. We also exclude medical-imaging AI, hospital administration tools, full-body scanning companies and drug-discovery platforms. ChatGPT and Claude are competitive threats, but they are broad technology platforms rather than AI doctor startups.

Funding figures below are approximate and reflect the latest publicly disclosed rounds. Private-company totals can differ slightly between announcements and financing databases.

Startup	What the startup currently does	Approximate cumulative funding
OpenEvidence	Gives physicians evidence-backed answers during clinical work	About $700 million
Hippocratic AI	Runs voice agents for follow-up, outreach and other non-diagnostic patient conversations	$404 million
K Health	Delivers AI-supported virtual care through major health systems	About $380 million
Ada Health	Assesses symptoms and guides patients toward appropriate care	About $167 million
Ubie	Operates consumer symptom-assessment and patient-navigation products	More than $125 million
Doctronic	Provides AI consultations, telehealth and autonomous prescription renewals in Utah	More than $65 million
Infermedica	Supplies symptom-checking and triage infrastructure to healthcare organizations	About $45 million
Lotus Health AI	Offers free AI-supported primary care with licensed physicians behind clinical actions	$41 million
Counsel Health	Combines medical AI with direct access to its own physicians	$36 million
UpDoc	Uses patient-facing AI to manage insulin inside clinician-set treatment plans	$18 million

Is there a clear AI doctor leader today?

K Health is ahead in the AI doctor market today, but OpenEvidence, Doctronic and Hippocratic AI each lead a major part of the category.

K Health has the broadest working care model. Its technology collects symptoms, reads available medical information, prepares recommendations and hands the case to a licensed clinician who can diagnose and treat the patient. The system is already used inside established health systems rather than being limited to a small experimental clinic.

OpenEvidence has far more visible usage and revenue. Its product, however, remains an intelligence tool for doctors. It can shape a medical decision, but the physician still runs the consultation, examines the patient and takes responsibility for treatment.

Doctronic is pushing furthest toward autonomy. According to the company’s Utah service information, its program covers around 190 eligible medications and charges patients $4 for a renewal. That is genuine independent medical action, but only inside a controlled program for people already taking maintenance medication.

Hippocratic AI leads patient conversation volume. The company currently reports more than 200 million interactions across over 1,000 use cases. Its agents do not diagnose or prescribe, making the product closer to an AI nurse, care coordinator or clinical call-center worker than a full AI doctor.

K Health leads complete AI-supported care. OpenEvidence leads clinical knowledge, Doctronic leads limited medical autonomy, and Hippocratic AI leads high-volume patient conversations.

If you want more recent data on this point, please see our latest digital health market report.

Which AI doctor startup is closest to replacing a real doctor visit?

K Health currently comes closest to replacing most of a routine virtual doctor visit, even though a licensed clinician still approves the final care decision.

The patient does not simply receive a list of possible conditions. K Health’s system asks follow-up questions, reviews records when available, organizes the case and prepares clinical recommendations. A doctor then checks the work, speaks with the patient when necessary and signs off on treatment.

K Health’s own partnership announcements show the model being used by Cedars-Sinai, Hartford HealthCare, Mass General Brigham, Northwell Health and Hackensack Meridian Health. More recently, Penn Medicine began working with K Health on patient-intake agents, Hartford expanded the partnership into endocrinology, and Novant Health introduced an integrated 24-hour primary-care model.

The repeated expansions are more convincing than any one pilot. Existing customers are giving K Health additional clinical work after seeing the product operate.

Counsel Health follows a similar path. Its AI gathers information and proposes next steps, while company-employed physicians remain responsible for care. The Wall Street Journal reported that Counsel had already reached more than 100,000 people through employers and insurers. K Health still has deeper hospital integrations and a longer record of repeat deployments.

Lotus Health AI offers a broader promise on paper: free primary care, prescriptions, lab orders and referrals. The service is active, and licensed clinicians remain accountable for clinical actions. Lotus has not disclosed enough patient volume, retention or treatment data to place it near K Health yet.

Doctronic is more independent when it renews an eligible Utah prescription. Across a full consultation, K Health covers much more of the doctor’s work.

Which AI doctor startup has built the biggest business?

OpenEvidence has built the biggest visible business in this comparison, although it sells intelligence to doctors rather than medical care to patients.

In its latest financing announcement, OpenEvidence reported more than $100 million in annualized revenue after building its commercial operation in less than a year. The company said usage had risen from about 3 million clinical consultations per month to 18 million within one year. It now reports more than one million consultations per day, equal to roughly 30 million in a 30-day month.

The scale gap is large. Doctronic reported eight-figure annualized revenue, meaning at least $10 million but less than $100 million. Even near the top of that range, OpenEvidence would remain larger. A pretty clear gap. K Health may generate substantial revenue through hospitals, but it does not publish a comparable current figure.

OpenEvidence’s free, advertising-supported product removes the subscription decision for doctors and helps explain its rapid spread. Hippocratic AI reports extraordinary activity but does not reveal enough revenue, contract value or usage depth to compare the business directly.

Startup	Best disclosed commercial evidence	What we can reasonably conclude
OpenEvidence	More than $100 million in annualized revenue and over one million daily physician consultations	Clear business leader by visible revenue and professional usage
Doctronic	Eight-figure annualized revenue after fifteenfold growth	Fast commercial challenger, still several times smaller
K Health	Repeated deployments and expansions across large health systems	Deep commercial use, but revenue and total encounter volume remain private
Hippocratic AI	More than 200 million patient interactions and over 60 partners	Huge operational reach, with unclear revenue per interaction
Counsel Health	More than 100,000 covered users	Real early distribution, well below the leaders
Lotus Health AI	Active free-care service backed by $41 million	Product is available, but commercial sustainability is unproven

If you want more recent data on this point, please see our latest digital health market report.

Which AI doctor startup is gaining ground fastest now?

Doctronic is growing fastest from a smaller base, while OpenEvidence is adding the most meaningful volume.

Doctronic said in its latest funding announcement that annualized revenue grew fifteenfold in less than six months, its repeat-patient rate nearly tripled and weekly unique visitors passed 300,000. These are company-reported figures rather than audited accounts, but revenue, repeat use and traffic all moved in the same direction.

Its regulatory progress makes that growth more important. Doctronic began as a free AI consultation product linked to ordinary telehealth. Utah’s AI Learning Lab then allowed it to renew eligible prescriptions without a physician checking every individual request.

OpenEvidence is adding far more users in absolute terms. Moving from 3 million to 18 million monthly consultations meant adding about 15 million monthly uses in one year. Its newer daily figure suggests adoption has continued rather than flattening after the initial surge.

K Health has lately shown a different kind of momentum. Hartford added endocrinology, Mass General Brigham widened access, Penn Medicine selected K Health for clinical AI work, and Novant launched another integrated primary-care service. Expansion by existing buyers gives more confidence than a long list of first-time pilots.

UpDoc also deserves attention. The FDA cleared its insulin-management system, and the American Diabetes Association’s Innovation Fund recently invested in the company. UpDoc is still entering pilots, but the clearance moved it into serious clinical competition.

Which AI doctor startup has the most mature clinical product and evidence?

K Health has the most mature combination of real clinical use and published evidence, while Ada Health has the strongest recent evidence for AI-guided care navigation.

K Health’s most useful study reviewed 461 real virtual urgent-care visits involving respiratory, urinary, vaginal, eye and dental problems. Medical experts rated the AI and physician recommendations as equal in 68% of cases, preferred the AI in 21% and preferred the physician in 11%. The cases came from Cedars-Sinai Connect rather than from fictional medical exam questions.

The study has limits. It covered common virtual-care conditions rather than medicine as a whole, and reviewers knew which recommendation came from the AI. Still, it tests a commercially useful question: can the system prepare a safe and sensible treatment recommendation inside a real patient workflow?

Ada’s newer study followed 1,470 adults using symptom assessment inside the app of CUF, a large Portuguese healthcare network. According to Ada’s publication of the results, one in three participants changed the level of care they intended to seek after using the tool. The share who remained unsure about what to do fell from 12.6% to 5%.

The Ada study is more useful as evidence of changed patient behavior than diagnostic superiority. AI-guided symptom assessment can alter what people decide to do when it connects directly to available care.

UpDoc has the strongest regulatory evidence for a tightly defined task. FDA documentation shows that clinicians set the insulin plan, dosing limits, glucose targets and safety rules. UpDoc then communicates with the patient and calculates instructions within those boundaries.

Doctronic’s retrospective analysis reported an 81% match with the leading diagnosis and 99.2% agreement on treatment plans across 500 telehealth cases. The results remain less convincing because the study was a preprint and included company-linked researchers.

Which AI doctor performs best on medical decisions today?

K Health has the strongest evidence for complete routine-care decisions, while OpenEvidence performs a different job: finding and organizing medical evidence for physicians.

K Health’s system takes a patient history, uses available records and proposes both a diagnosis and a treatment plan. Its published comparison found the AI recommendation was rated at least as good as the physician’s in 89% of reviewed cases when ties and AI wins were combined.

That percentage should not be stretched into a claim that K Health is better than doctors across medicine. It shows the product can contribute useful decisions in common virtual urgent-care cases.

OpenEvidence faces a more complicated picture. A Nature Medicine evaluation compared OpenEvidence and UpToDate Expert AI with GPT-5.2, Gemini 3.1 Pro and Claude Opus 4.6. The general-purpose frontier models performed better across the study’s public benchmarks and questions submitted by physicians.

The result weakens the assumption that a specialist medical company automatically owns the best medical model. Doctors may still prefer OpenEvidence because it retrieves licensed literature, displays sources clearly and fits the way they already research clinical questions.

OpenEvidence has lately strengthened that evidence layer through content relationships involving NEJM, JAMA, NCCN, Wiley and Cochrane. It also introduced EvidenceGrade, which shows physicians how strong the evidence behind an answer is.

K Health currently has the better evidence on patient-level treatment decisions. OpenEvidence has the stronger product for medical research and source verification.

If you want more recent data on this point, please see our latest digital health market report.

Which AI doctor has the best economics for patients and healthcare buyers?

Doctronic currently offers the clearest direct saving for patients, while OpenEvidence has the strongest model for rapid professional adoption.

A Utah patient can request an eligible prescription renewal for $4 and receive it through the normal pharmacy system. A traditional appointment can cost many times more and may require days of waiting. The service is narrow, but the saving is immediate.

Lotus Health AI goes further by charging patients nothing for primary-care consultations, prescriptions, lab orders or referrals. The company says investor capital and advertising currently support the service. Free care may attract users quickly, although Lotus has not yet shown that the model can cover medical staffing, compliance and clinical operations over time.

Counsel has a more conventional offer. Its direct consumer option has been priced at $29 for a week of physician access, while employers and insurers can pay to cover larger populations. Counsel has said its doctors can manage as many as 15 patients per hour with AI support. That figure comes from the company and still needs independent confirmation.

OpenEvidence removes the price barrier for physicians entirely. The free, advertising-supported model allowed it to spread without waiting for every hospital to approve a new software budget. It is currently the category’s strongest distribution model, although advertising near medical information will continue to raise questions.

K Health may create more total value for a health system by reducing intake work and preparing cases before the physician enters. Contract values, implementation costs and verified savings remain private, preventing a clean economic comparison.

Which AI doctor startup has proved it can scale with serious healthcare customers?

K Health has done the best job of scaling complete clinical care with major healthcare customers, while OpenEvidence and Hippocratic AI lead in software interaction volume.

OpenEvidence can add another physician consultation without adding a clinician for every answer. Hippocratic AI has the same advantage in voice-based follow-up, outreach and education. Its reported interaction volume proves the infrastructure can handle patient conversations at exceptional scale.

Complete medical care is harder to expand. Liability rises, medical records must stay accurate, local rules differ and unusual cases need a qualified professional. K Health has built around these constraints by integrating with health systems and keeping licensed clinicians responsible for treatment.

Its customer list includes Cedars-Sinai, Hartford HealthCare, Mass General Brigham, Northwell Health, Hackensack Meridian Health and Penn Medicine. Several relationships have expanded after launch, which tells us more than a collection of innovation announcements.

Hartford added endocrinology after beginning with general virtual care. Mass General Brigham widened access. Penn Medicine is using K Health technology for patient intake. Novant Health introduced an integrated 24-hour primary-care model.

OpenEvidence has built the strongest network around physician knowledge. Sutter Health is bringing the product into clinical workflows, while NewYork-Presbyterian, Columbia and Weill Cornell have expanded access across their organizations. Its publishing and medical-association relationships also give doctors access to material that general chatbots may not carry.

Doctronic has proved its autonomous renewal model can operate under one state program. UpDoc has FDA clearance and has announced pilots with organizations including Cleveland Clinic, Highmark Health, Stanford Health Care and UCSF Health. Neither company has yet shown broad, repeatable deployment across multiple healthcare systems.

K Health currently wins this comparison because its customers are trusting the product with more of the actual care process.

What can the leading AI doctor startups do that competitors cannot easily copy?

K Health currently has the hardest overall position to reproduce because its advantage comes from hospital access, clinical operations and workflow integration.

A competitor can build a medical chatbot much faster than it can secure health-system contracts, pass security reviews, connect patient records, recruit licensed clinicians and adapt a product to hospital protocols. K Health already combines those pieces.

OpenEvidence has another strong position. Its verified-doctor user base produces a large stream of real clinical questions, while licensed content and medical partnerships strengthen the answers. Frontier models remain a threat because OpenAI and Anthropic can improve the underlying intelligence faster than most healthcare startups.

Doctronic is accumulating regulatory experience through Utah’s AI Learning Lab. That knowledge becomes valuable if similar autonomous-care programs open elsewhere.

Hippocratic AI has learned from a very large number of patient conversations, which may improve voice quality, safety and the handling of difficult exchanges. Since its agents avoid diagnosis and prescribing, that advantage currently protects patient communication rather than full medical care.

K Health’s moat is the broadest today. OpenEvidence has the strongest professional network, while Doctronic and Hippocratic AI hold narrower advantages that could become more valuable as their markets expand.

If you want more recent data on this point, please see our latest digital health market report.

Is the best-funded AI doctor startup actually using its money well?

OpenEvidence is currently converting large amounts of funding into adoption more effectively than any other heavily funded company in the field.

The company raised roughly $700 million over twelve months while passing $100 million in annualized revenue and continuing to grow physician use. The valuation may be aggressive, but the financing supports an already large product rather than a distant promise.

Doctronic looks strongest on capital efficiency. With more than $65 million raised, it reached eight-figure annualized revenue, attracted more than 300,000 weekly visitors and launched an active autonomous prescription-renewal program. The company reported those figures itself, but the output remains substantial for less than one-tenth of OpenEvidence’s funding.

K Health has needed much more capital because it is building software and a care-delivery operation. Its move away from a standalone consumer subscription product shows that an earlier route did not become the final strategy. The hospital-focused model gives K Health access to existing patients, records and trusted medical brands.

Hippocratic AI has turned $404 million into vast interaction volume, international partnerships and an advanced voice platform. We still lack enough contract and revenue information to judge whether that scale is financially efficient.

UpDoc is the most interesting small outlier. It reached FDA clearance with $18 million in funding and added the American Diabetes Association’s Innovation Fund as an investor. Commercial scale remains limited, but the regulatory progress per dollar is unusually strong. Still early, though.

OpenEvidence is using the most money effectively. Doctronic appears to use money most efficiently, while UpDoc has achieved the strongest regulatory progress with a small funding base.

Is one startup leading the whole AI doctor market?

K Health covers more of the complete AI doctor pathway than any rival, although different companies still lead individual parts of care.

Ada, Ubie and Infermedica are strongest when patients first assess symptoms. OpenEvidence leads physician research. Hippocratic AI leads repetitive patient conversations. UpDoc manages a narrow part of chronic care, while Doctronic is furthest ahead in autonomous medical action.

K Health spans more of the pathway. It collects patient information, works with medical records, prepares recommendations and supports treatment inside health systems. A clinician remains responsible, but the AI handles a substantial share of the surrounding work.

We rank K Health first even though OpenEvidence is larger and Doctronic is more autonomous. The ranking could change if Doctronic expands beyond prescription renewals, OpenEvidence begins managing care directly or UpDoc proves its model across large patient populations.

Which AI doctor startups are actually ahead?

K Health is the AI doctor startup ahead overall today, with OpenEvidence close behind in strategic importance and Doctronic emerging as the most dangerous challenger.

K Health wins because it has the strongest mix of real care delivery, hospital distribution, product maturity and clinical evidence. It already works inside major healthcare organizations, and several customers are expanding how they use it. Its lead is meaningful, but the company has not pulled far enough ahead to make the race predictable.

OpenEvidence ranks second because no other medical AI startup comes close to its physician usage, visible revenue or knowledge ecosystem. It would rank first in a comparison of AI tools for doctors. The product does not currently deliver the consultation itself, which keeps it behind K Health here.

Doctronic ranks third and has the clearest route upward. Its growth, repeat use and Utah prescription program show that patients will accept meaningful medical action from AI. Broader regulatory approval and stronger independent evidence could move Doctronic into the top two.

Hippocratic AI ranks fourth because its patient reach is far larger than most competitors. Its refusal to diagnose or prescribe limits how much of a doctor’s work it can currently perform.

Ada remains the strongest mature symptom-assessment company. Counsel is the most credible younger physician-supervised care challenger. UpDoc has stronger regulatory progress but a much narrower product. Ubie and Infermedica have broad reach in navigation and triage, while Lotus still needs to prove that its free-care model can attract and retain patients at scale.

Rank	Startup	Why the startup holds this position
1	K Health	The most complete AI-supported care platform, with strong health-system distribution, real clinical workflows and the best evidence on full treatment recommendations
2	OpenEvidence	The clear leader in physician usage, visible revenue and medical knowledge infrastructure, but still a tool for doctors rather than a doctor service
3	Doctronic	The fastest-growing challenger and the current leader in limited autonomous medical action
4	Hippocratic AI	Far ahead in patient conversation volume, although its agents still avoid diagnosis and prescribing
5	Ada Health	The most mature symptom-assessment specialist, with regulated products and newer real-world evidence showing changes in patient behavior
6	Counsel Health	A credible physician-supervised model with real employer and insurer distribution
7	UpDoc	The strongest narrow regulatory achievement, with FDA-cleared insulin management inside clinician-set limits
8	Ubie	A large consumer and provider-navigation platform, but still far from delivering complete care
9	Infermedica	Established international triage infrastructure with less visible recent momentum
10	Lotus Health AI	A bold free-care product whose usage, clinical outcomes and long-term economics remain largely undisclosed

If you want more recent data on this point, please see our latest digital health market report.

Who is the author of this content?

NEW MARKET PITCH TEAM

We track new markets so founders and investors can move faster

We build living "market pitch" documents for emerging markets: AI, synthetic biology, new proteins, and more. Instead of outdated PDFs or hallucinated LLM answers, our clients get a clean, visual, always-updated view of what's really happening: key players, deals, regulations, and signals that matter. Learn more about us.